. clear all
. cap noi which tabout
c:\ado\plus\t\tabout.ado
*! 2.0.8 Ian Watson 15mar2019
*! tabout version 3 (beta) available at: http://tabout.net.au
. if _rc==111 {
. cap noi ssc install tabout
. }
. cap noi which pathutil
c:\ado\plus\p\pathutil.ado
*! version 2.2.0 19nov2020 daniel klein
. if _rc==111 {
. cap noi net install pathutil, from("http://fmwww.bc.edu/repec/bocode/p/")
. }
. cap noi which pathutil
c:\ado\plus\p\pathutil.ado
*! version 2.2.0 19nov2020 daniel klein
. if _rc==111 {
. ssc install dirtools
. }
. cap noi which project
c:\ado\plus\p\project.ado
*! version 1.3.1 22dec2013 picard@netbox.com
. if _rc==111 {
. ssc install project
. }
. cap noi which stipw
c:\ado\plus\s\stipw.ado
*! Version 1.0.0 17Jan2022
. if _rc==111 {
. ssc install stipw
. }
. cap noi which stpm2
c:\ado\plus\s\stpm2.ado
*! version 1.7.5 May2021
. if _rc==111 {
. ssc install stpm2
. }
. cap noi which rcsgen
c:\ado\plus\r\rcsgen.ado
*! version 1.5.9 13FEB2022
. if _rc==111 {
. ssc install rcsgen
. }
. cap noi which matselrc
c:\ado\plus\m\matselrc.ado
*! NJC 1.1.0 20 Apr 2000 (STB-56: dm79)
. if _rc==111 {
. cap noi net install dm79, from(http://www.stata.com/stb/stb56)
. }
Date created: 14:53:35 16 Feb 2023.
Get the folder
E:\Mi unidad\Alvacast\SISTRAT 2022 (github)
Fecha: 16 Feb 2023, considerando un SO Windows para el usuario: andre
Path data= ;
Time: 16 Feb 2023, considering an OS Windows
The file is located and named as: E:\Mi unidad\Alvacast\SISTRAT 2022 (github)fiscalia_mariel_oct_2022_match_SENDA_pris.dta
=============================================================================
=============================================================================
We open the files
. use "fiscalia_mariel_oct_2022_match_SENDA_pris.dta", clear
. encode escolaridad_rec, gen(esc_rec)
. encode sex, generate(sex_enc)
. encode sus_principal_mod, generate(sus_prin_mod)
. encode freq_cons_sus_prin, generate(fr_sus_prin)
. encode compromiso_biopsicosocial, generate(comp_biosoc)
. encode tenencia_de_la_vivienda_mod, generate(ten_viv)
. *encode dg_cie_10_rec, generate(dg_cie_10_mental_h) *already numeric
. encode dg_trs_cons_sus_or, generate(sud_severity_icd10)
. encode macrozona, generate(macrozone)
. gen motivodeegreso_mod_imp_rec3 = 1
. replace motivodeegreso_mod_imp_rec3 = 2 if strpos(motivodeegreso_mod_imp_rec,"Early")>0
(15,797 real changes made)
. replace motivodeegreso_mod_imp_rec3 = 3 if strpos(motivodeegreso_mod_imp_rec,"Late")>0
(35,781 real changes made)
.
. *encode policonsumo, generate(policon) *already numeric
.
. *motivodeegreso_mod_imp_rec3 edad_al_ing_1 edad_ini_cons dias_treat_imp_sin_na_1 i.escolaridad_rec i.sus_principal_mod i.freq
> _cons_sus_prin i.compromiso_biopsicosocial i.tenencia_de_la_vivienda_mod i.dg_cie_10_rec i.dg_trs_cons_sus_or i.macrozona i.n
> _off_vio i.n_off_acq i.n_off_sud i.n_off_oth
Then we set the data base in surirval format and bring the urban-rural classification of municipallities from this link.
. cap qui noi frame create temp
. frame temp: import excel "Clasificacion-comunas-PNDR.xlsx", firstrow clear
(11 vars, 345 obs)
. *frame temp: browse
. frame change default
.
. *select code of municipality
. gen str20 comuna = ustrregexs(1) if ustrregexm(comuna_residencia_cod,"([\d,]+)")
(2 missing values generated)
.
. *recode comuna if
. *http://www.sinim.cl/archivos/centro_descargas/modificacion_instructivo_pres_codigos.pdf
. *file:///C:/Users/CISSFO~1/AppData/Local/Temp/MicrosoftEdgeDownloads/4ef08de9-6832-4db6-8124-f69a7b256270/codigoComunas-20180
> 801%20(1).pdf
.
. replace comuna= "16101" if strpos(strlower(comuna),"8401")>0
(434 real changes made)
. replace comuna= "16102" if strpos(strlower(comuna),"8402")>0
(10 real changes made)
. replace comuna= "16103" if strpos(strlower(comuna),"8406")>0
(32 real changes made)
. replace comuna= "16104" if strpos(strlower(comuna),"8407")>0
(2 real changes made)
. replace comuna= "16105" if strpos(strlower(comuna),"8410")>0
(1 real change made)
. replace comuna= "16106" if strpos(strlower(comuna),"8411")>0
(8 real changes made)
. replace comuna= "16107" if strpos(strlower(comuna),"8413")>0
(12 real changes made)
. replace comuna= "16108" if strpos(strlower(comuna),"8418")>0
(4 real changes made)
. replace comuna= "16109" if strpos(strlower(comuna),"8421")>0
(6 real changes made)
. replace comuna= "16201" if strpos(strlower(comuna),"8414")>0
(22 real changes made)
. replace comuna= "16202" if strpos(strlower(comuna),"8403")>0
(0 real changes made)
. replace comuna= "16203" if strpos(strlower(comuna),"8404")>0
(13 real changes made)
. replace comuna= "16204" if strpos(strlower(comuna),"8408")>0
(1 real change made)
. replace comuna= "16205" if strpos(strlower(comuna),"8412")>0
(1 real change made)
. replace comuna= "16206" if strpos(strlower(comuna),"8415")>0
(2 real changes made)
. replace comuna= "16207" if strpos(strlower(comuna),"8420")>0
(1 real change made)
. replace comuna= "16301" if strpos(strlower(comuna),"8416")>0
(12 real changes made)
. replace comuna= "16302" if strpos(strlower(comuna),"8405")>0
(6 real changes made)
. replace comuna= "16303" if strpos(strlower(comuna),"8409")>0
(0 real changes made)
. replace comuna= "16304" if strpos(strlower(comuna),"8417")>0
(0 real changes made)
. replace comuna= "16305" if strpos(strlower(comuna),"8419")>0
(0 real changes made)
.
. destring comuna, replace
comuna: all characters numeric; replaced as int
(2 missing values generated)
.
. *frame temp: gen str20 comuna = ustrregexs(1) if ustrregexm(cod_com,"([\d,]+)")
.
. frlink m:1 comuna, frame(temp cod_com) //*Clasificación
(2 observations in frame default unmatched)
. frget Clasificación, from(temp)
(2 missing values generated)
(1 variable copied from linked frame)
.
. encode Clasificación, generate(clas)
. *70,863
. *si no está perdido cod_region, significa que hubo un registro (0/1) y el tiempo es el tiempo desde
. *set the indicator
. gen event=0
. replace event=1 if !missing(offender_d)
(5,144 real changes made)
. *replace event=1 if !missing(sex)
.
. gen diff= age_offending_imp-edad_al_egres_imp
.
. *age time
. stset age_offending_imp, fail(event ==1) enter(edad_al_egres_imp)
failure event: event == 1
obs. time interval: (0, age_offending_imp]
enter on or after: time edad_al_egres_imp
exit on or before: failure
------------------------------------------------------------------------------
70,863 total observations
1 observation ends on or before enter()
------------------------------------------------------------------------------
70,862 observations remaining, representing
5,144 failures in single-record/single-failure data
302,812.78 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 10.95068
last observed exit t = 90.65027
.
. stdescribe, weight
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
|-------------- per subject --------------|
Category total mean min median max
------------------------------------------------------------------------------
no. of subjects 70862
no. of records 70862 1 1 1 1
(first) entry time 36.52205 10.95068 34.6274 88.91507
(final) exit time 40.79532 14.98082 39.0765 90.65027
subjects with gap 0
time on gap if gap 0
time at risk 302812.78 4.273275 .0000449 3.964384 10.75828
failures 5144 .0725918 0 0 1
------------------------------------------------------------------------------
We calculate the incidence rate.
. stsum, by (motivodeegreso_mod_imp_rec)
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
| Incidence Number of |------ Survival time -----|
motivo~c | Time at risk rate subjects 25% 50% 75%
---------+---------------------------------------------------------------------
Treatmen | 76,631.0344 .0086649 19275 34.88843 . .
Treatmen | 65,879.5067 .0259717 15797 23.57016 34.91855 .
Treatmen | 160,259.188 .0172595 35781 25.36071 48.78303 .
---------+---------------------------------------------------------------------
Total | 302,769.729 .0169799 70853 24.87337 46.45585 .
We open the files
. *si no está perdido cod_region, significa que hubo un registro (0/1) y el tiempo es el tiempo desde
. *set the indicator
. gen event=0
variable event already defined
r(110);
. replace event=1 if !missing(offender_d)
(0 real changes made)
. *replace event=1 if !missing(sex)
.
. gen diff= age_offending_imp-edad_al_egres_imp
variable diff already defined
r(110);
.
. *age time
. stset age_offending_imp, fail(event ==1) enter(edad_al_egres_imp)
failure event: event == 1
obs. time interval: (0, age_offending_imp]
enter on or after: time edad_al_egres_imp
exit on or before: failure
------------------------------------------------------------------------------
70,863 total observations
1 observation ends on or before enter()
------------------------------------------------------------------------------
70,862 observations remaining, representing
5,144 failures in single-record/single-failure data
302,812.78 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 10.95068
last observed exit t = 90.65027
.
. stdescribe, weight
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
|-------------- per subject --------------|
Category total mean min median max
------------------------------------------------------------------------------
no. of subjects 70862
no. of records 70862 1 1 1 1
(first) entry time 36.52205 10.95068 34.6274 88.91507
(final) exit time 40.79532 14.98082 39.0765 90.65027
subjects with gap 0
time on gap if gap 0
time at risk 302812.78 4.273275 .0000449 3.964384 10.75828
failures 5144 .0725918 0 0 1
------------------------------------------------------------------------------
We calculate the incidence rate.
. stsum, by (motivodeegreso_mod_imp_rec)
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
| Incidence Number of |------ Survival time -----|
motivo~c | Time at risk rate subjects 25% 50% 75%
---------+---------------------------------------------------------------------
Treatmen | 76,631.0344 .0086649 19275 34.88843 . .
Treatmen | 65,879.5067 .0259717 15797 23.57016 34.91855 .
Treatmen | 160,259.188 .0172595 35781 25.36071 48.78303 .
---------+---------------------------------------------------------------------
Total | 302,769.729 .0169799 70853 24.87337 46.45585 .
=============================================================================
=============================================================================
.
. local ttl `" "Tr Completion" "Tr Disch (Early)" "Tr Disch (Late)" "'
. forvalues i = 1/3 {
2. cap drop sum_*
3. gettoken title ttl: ttl
4. cap qui egen sum_`i' = total(_t) if motivodeegreso_mod_imp_rec3==`i'
5. cap qui tab _d motivodeegreso_mod_imp_rec3 if _d==1 & motivodeegreso_mod_imp_rec3==`i'
6. scalar n_eventos`i' =r(N)
7. qui tabstat sum_`i', save
8. scalar sum_`i'_total =r(StatTotal)[1,1]
9. scalar ir_t = round((n_eventos`i'/sum_`i'_total)*1000,.1)
10. *di ir_t
. *es numérico, para no generar incompatibilidad, se pasa a escalar
. global dens_inc`i' "Incidence rate ratio for `title' (`i') (x1,000 person-days): `=scalar(ir_t)'"
11. *di %9.1f (n_eventos`i'/sum_`i'_total)*1000
. }
Incidence rate ratio for Tr Completion (1) (x1,000 person-days): .8
Incidence rate ratio for Tr Disch (Early) (2) (x1,000 person-days): 2.9
Incidence rate ratio for Tr Disch (Late) (3) (x1,000 person-days): 1.9
We recode the discharge cause to contrast them
. cap noi recode motivodeegreso_mod_imp_rec3 (1=0 "Tr Completion" ) ///
> (2=1 "Early Disch") ///
> (else=.), gen(tto_2_1)
(70863 differences between motivodeegreso_mod_imp_rec3 and tto_2_1)
. cap noi recode motivodeegreso_mod_imp_rec3 (1=0 "Tr Completion" ) ///
> (3=1 "Late Disch") ///
> (else=.), gen(tto_3_1)
(70863 differences between motivodeegreso_mod_imp_rec3 and tto_3_1)
. cap noi recode motivodeegreso_mod_imp_rec3 (2=0 "Early Disch" ) ///
> (3=1 "Late Disch") ///
> (else=.), gen(tto_3_2)
(70863 differences between motivodeegreso_mod_imp_rec3 and tto_3_2)
We explored the inicidence rate ratios (IRR) of each cause of discharge.
.
. *set trace on
. local stname `" "2_1" "3_1" "3_2" "'
. local titl `" "Early Disch vs Tr Completion" "Late Disch vs Tr Completion" "Late vs Early Disch" "'
. foreach s of local stname {
2. gettoken title titl: titl
3. cap noi qui ir _d tto_`s' _t
4. scalar ir_`s' =round(r(irr),.01)
5. *di ir_`s'
. scalar ir_`s'_lb =round(r(lb_irr),.01)
6. *di ir_`s'_lb
. scalar ir_`s'_ub =round(r(ub_irr),.01)
7. *di ir_`s'_ub
. local ir1= ir_`s'
8. local ir2= ir_`s'_lb
9. local ir3= ir_`s'_ub
10. *di in gr _col(13) " `title': IRR `ir1' (IC 95% `ir2' - `ir3') "
. global irr_`s' " `title': IRR `ir1' (IC 95% `ir2' - `ir3') "
11. global ir_`s' "`ir1' (IC 95% `ir2' - `ir3')"
12. }
We generated a log-rank test to compare the expected to the observed result, obtaining that:
Early Disch vs Tr Completion: IRR 3.62 (IC 95% 3.31 - 3.96) , so patients with an Early Discharge had a greater incidence rate than patients in Tr Completion
Late Disch vs Tr Completion: IRR 2.43 (IC 95% 2.23 - 2.65) , so patients with a Late Discharge had a greater incidence rate than patients in Tr Completion
Late vs Early Disch: IRR .67 (IC 95% .63 - .71) , so patients with a Late Discharge had a lower incidence rate than patients with an Early Discharge
. *set trace on
. local stname `" "2_1" "3_1" "3_2" "'
. local titl `" "Early Disch vs Tr Completion" "Late Disch vs Tr Completion" "Late vs Early Disch" "'
. foreach s of local stname {
2. gettoken title titl: titl
3. cap noi qui sts test tto_`s', logrank
4. scalar logrank_chi`s'= round(r(chi2),.01)
5. scalar logrank_df`s'= r(df)
6. scalar logrank_p_`s'= round(chiprob(r(df),r(chi2)),.001)
7. local lr1= logrank_chi`s'
8. local lr2= logrank_df`s'
9. local lr3= logrank_p_`s'
10. global logrank_`s' " Chi^2(`lr2')=`lr1',p=`lr3'"
11. }
. *set trace off
With a value of Chi^2(1)=328.66,p=0, survival curves between patients that had an Early Disch y Tr Completion were significantly different.
With a value of Chi^2(1)=142.23,p=0, survival curves between patients that had an Late Disch y Tr Completion were significantly different.
With a value of Chi^2(1)=97.22,p=0, survival curves between patients that had an Late y Early Disch were significantly different.
=============================================================================
=============================================================================
We generated a graph with every type of treatment and the Nelson-Aalen estimate.
. cap rm "tto2.svg"
. sts graph, na by (motivodeegreso_mod_imp_rec) ci ///
> title("Comission of an offense (end with imprisonment)") ///
> subtitle("Nelson-Aalen Cum Hazards w/ Confidence Intervals 95%") ///
> risktable(, size(*.5) order(1 "Tr Completion" 2 "Early Disch" 3 "Late Disch")) ///
> ytitle("Cum. Hazards") ylabel(#8) ///
> xtitle("Years of age") xlabel(#8) ///
> note("Source: nDP, SENDA's SUD Treatments & POs Office Data period 2010-2019 ") ///
> legend(rows(3)) ///
> legend(cols(4)) ///
> graphregion(color(white) lwidth(large)) bgcolor(white) ///
> plotregion(fcolor(white)) graphregion(fcolor(white) ) /// //text(.5 1 "IR = <0.001") ///
> legend(order(1 "95CI Tr Completion" 2 "Tr Completion" 3 "95CI Early Tr Disch" 4 "Early Tr Disch " 5 "95CI Late Tr Disch" 6 "L
> ate Tr Disch" )size(*.5)region(lstyle(none)) region(c(none)) nobox)
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
(note: named style large not found in class linewidth, default attributes used)
(note: linewidth not found in scheme, default attributes used)
(note: named style large not found in class linewidth, default attributes used)
(note: linewidth not found in scheme, default attributes used)
. graph save "`c(pwd)'\_figs\tto2.gph", replace
(note: file E:\Mi unidad\Alvacast\SISTRAT 2022 (github)\_figs\tto2.gph not found)
(file E:\Mi unidad\Alvacast\SISTRAT 2022 (github)\_figs\tto2.gph saved)
=============================================================================
=============================================================================
We tested the schoefeld residuals.
. *c("edad_al_ing_1", "edad_ini_cons", "dias_treat_imp_sin_na_1", "escolaridad_rec", "sus_principal_mod", "freq_cons_sus_prin",
> "compromiso_biopsicosocial", "tenencia_de_la_vivienda_mod", "dg_cie_10_rec", "dg_trs_cons_sus_or", "macrozona", "policonsumo
> ", "n_prev_off", "n_off_vio", "n_off_acq", "n_off_sud", "n_off_oth")
.
. global sim 1e5 //5e1 1e5
. global boots 1e3 //5e1 2e3
. global times 0 90 365 1096 1826
. range timevar0 90 1826 90
(70,773 missing values generated)
.
. global covs "edad_al_ing_1 edad_ini_cons dias_treat_imp_sin_na_1 sex esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten_viv dg_
> cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas"
. global covs_2 "motivodeegreso_mod_imp_rec3 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc t
> en_viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas"
.
.
. stcox $covs_2 , efron robust nolog schoenfeld(sch*) scaledsch(sca*)
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
Cox regression -- Efron method for ties
No. of subjects = 59,755 Number of obs = 59,755
No. of failures = 3,947
Time at risk = 233401.5585
Wald chi2(17) = 2899.48
Log pseudolikelihood = -33229.301 Prob > chi2 = 0.0000
---------------------------------------------------------------------------------------------
| Robust
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
motivodeegreso_mod_imp_rec3 | 1.137893 .0237419 6.19 0.000 1.092298 1.185391
edad_al_ing_1 | 1.173086 .0118876 15.75 0.000 1.150016 1.196618
edad_ini_cons | .9763696 .004763 -4.90 0.000 .9670788 .9857497
sex_enc | .6071509 .0270164 -11.21 0.000 .5564431 .6624796
esc_rec | 1.298227 .033886 10.00 0.000 1.233482 1.366371
sus_prin_mod | 1.253118 .0214601 13.18 0.000 1.211755 1.295893
fr_sus_prin | 1.033203 .0171907 1.96 0.050 1.000053 1.067452
comp_biosoc | 1.255332 .0380386 7.50 0.000 1.182948 1.332144
ten_viv | .9701535 .0155257 -1.89 0.058 .940196 1.001066
dg_cie_10_rec | 1.042133 .01934 2.22 0.026 1.004908 1.080736
sud_severity_icd10 | .9046789 .0384707 -2.36 0.018 .8323343 .9833115
macrozone | 1.273756 .0305628 10.08 0.000 1.215241 1.335089
policonsumo | 1.10135 .0537876 1.98 0.048 1.000817 1.211982
n_off_vio | 1.455552 .0548066 9.97 0.000 1.352001 1.567034
n_off_acq | 2.888106 .0998208 30.69 0.000 2.69894 3.09053
n_off_sud | 1.342254 .0498049 7.93 0.000 1.248103 1.443507
clas | 1.127385 .0331957 4.07 0.000 1.064165 1.194362
---------------------------------------------------------------------------------------------
. estat phtest, log detail
Test of proportional-hazards assumption
Time: Log(t)
----------------------------------------------------------------
| rho chi2 df Prob>chi2
------------+---------------------------------------------------
motivodeeg~3| 0.00439 0.06 1 0.8000
edad_al_in~1| -0.03089 4.84 1 0.0279
edad_ini_c~s| 0.03017 4.18 1 0.0410
sex_enc | 0.03116 4.01 1 0.0452
esc_rec | -0.04704 8.21 1 0.0042
sus_prin_mod| -0.00195 0.01 1 0.9155
fr_sus_prin | -0.00758 0.23 1 0.6318
comp_biosoc | -0.02198 1.93 1 0.1646
ten_viv | -0.02151 1.99 1 0.1585
dg_cie_10_~c| -0.00416 0.07 1 0.7978
sud_sever~10| 0.02506 2.52 1 0.1124
macrozone | 0.00395 0.06 1 0.8121
policonsumo | 0.02191 1.98 1 0.1595
n_off_vio | 0.01895 1.60 1 0.2060
n_off_acq | 0.01445 0.97 1 0.3255
n_off_sud | 0.05507 13.68 1 0.0002
clas | -0.01318 0.77 1 0.3810
------------+---------------------------------------------------
global test | 47.44 17 0.0001
----------------------------------------------------------------
note: robust variance-covariance matrix used.
. scalar chi2_scho_test = r(chi2)
.
. mat mat_scho_test = r(phtest)
.
. esttab matrix(mat_scho_test) using "mat_scho_test2.csv", replace
(output written to mat_scho_test2.csv)
. esttab matrix(mat_scho_test) using "mat_scho_test2.html", replace
(output written to mat_scho_test2.html)
.
| mat_scho_test | ||||
| rho | chi2 | df | p | |
| motivodeegreso_mod_imp_rec3 | .0043948 | .0642029 | 1 | .7999724 |
| edad_al_ing_1 | -.0308941 | 4.835282 | 1 | .0278831 |
| edad_ini_cons | .0301658 | 4.177453 | 1 | .0409652 |
| sex_enc | .031157 | 4.012017 | 1 | .0451771 |
| esc_rec | -.0470359 | 8.208679 | 1 | .004169 |
| sus_prin_mod | -.0019499 | .0112656 | 1 | .9154718 |
| fr_sus_prin | -.0075836 | .2295747 | 1 | .6318394 |
| comp_biosoc | -.0219752 | 1.931193 | 1 | .164628 |
| ten_viv | -.0215142 | 1.988285 | 1 | .1585203 |
| dg_cie_10_rec | -.0041595 | .0656467 | 1 | .7977839 |
| sud_severity_icd10 | .0250598 | 2.52076 | 1 | .1123564 |
| macrozone | .0039451 | .0565365 | 1 | .8120563 |
| policonsumo | .0219067 | 1.978925 | 1 | .1595037 |
| n_off_vio | .0189453 | 1.599513 | 1 | .2059723 |
| n_off_acq | .0144483 | .9666242 | 1 | .3255235 |
| n_off_sud | .055072 | 13.68308 | 1 | .0002164 |
| clas | -.0131839 | .7673791 | 1 | .3810291 |
We generated a list of parametric survival models with different distributions (Exponential, Weibull, Gompertz, Log-logistic, Log-normal & Generalized gamma). Aditionally, we defined a series of Royston-Parmar models with a function of restricted cubic splines, in which the knots (#df -1) are defined in each percentile of the distribution. We saved the estimates in the file called `parmodels_m2_nov_22_2’.
.
. // Cox w/tvc
. forvalues j=1/7 {
2. di in yellow "{bf: ***********}"
3. di in yellow "{bf: family Cox tvc `j'}"
4. di in yellow "{bf: ***********}"
5. set seed 2125
6. qui cap noi stmerlin $covs_2 , dist(exponential) tvc(motivodeegreso_mod_imp_rec3) dftvc(`j')
7. estimates store m2_1_cox`j'
8. }
***********
family Cox tvc 1
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_1
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20468.693
Iteration 2: log likelihood = -18700.231
Iteration 3: log likelihood = -18114.538
Iteration 4: log likelihood = -18113.432
Iteration 5: log likelihood = -18113.432
Survival model Number of obs = 59,755
Log likelihood = -18113.432
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .0588047 .0229598 2.56 0.010 .0138043 .103805
edad_al_in~1 | -.0045602 .0042316 -1.08 0.281 -.0128541 .0037336
edad_ini_c~s | -.0255373 .0046225 -5.52 0.000 -.0345971 -.0164774
sex_enc | -.5072461 .0441771 -11.48 0.000 -.5938316 -.4206606
esc_rec | .2281343 .0263698 8.65 0.000 .1764505 .2798181
sus_prin_mod | .2179554 .0180669 12.06 0.000 .1825449 .2533658
fr_sus_prin | .0339021 .0166069 2.04 0.041 .0013532 .066451
comp_biosoc | .2212173 .0301702 7.33 0.000 .1620848 .2803497
ten_viv | -.017707 .0156754 -1.13 0.259 -.0484303 .0130163
dg_cie_10_~c | .03996 .0185405 2.16 0.031 .0036214 .0762986
sud_sever~10 | -.1088304 .0422921 -2.57 0.010 -.1917214 -.0259394
macrozone | .2369395 .0243234 9.74 0.000 .1892664 .2846125
policonsumo | .0886144 .0487491 1.82 0.069 -.0069321 .184161
n_off_vio | .4122978 .0365609 11.28 0.000 .3406397 .4839559
n_off_acq | 1.116647 .0337349 33.10 0.000 1.050527 1.182766
n_off_sud | .3488887 .0357897 9.75 0.000 .2787423 .4190352
clas | .1092927 .0286241 3.82 0.000 .0531904 .165395
motivodeeg~( | -.135411 .0156989 -8.63 0.000 -.1661802 -.1046417
_cons | -5.930007 .2478498 -23.93 0.000 -6.415784 -5.44423
------------------------------------------------------------------------------
***********
family Cox tvc 2
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_2
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20467.93
Iteration 2: log likelihood = -18716.062
Iteration 3: log likelihood = -18083.554
Iteration 4: log likelihood = -18079.751
Iteration 5: log likelihood = -18079.745
Iteration 6: log likelihood = -18079.745
Survival model Number of obs = 59,755
Log likelihood = -18079.745
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0118597 .0250325 -0.47 0.636 -.0609225 .0372032
edad_al_in~1 | .0132548 .004782 2.77 0.006 .0038823 .0226274
edad_ini_c~s | -.025945 .0046742 -5.55 0.000 -.0351063 -.0167837
sex_enc | -.5065669 .0441583 -11.47 0.000 -.5931156 -.4200181
esc_rec | .2478654 .0264747 9.36 0.000 .1959759 .2997548
sus_prin_mod | .2196178 .0181557 12.10 0.000 .1840333 .2552023
fr_sus_prin | .0367608 .0165879 2.22 0.027 .004249 .0692725
comp_biosoc | .2220271 .0301794 7.36 0.000 .1628765 .2811776
ten_viv | -.0234643 .015678 -1.50 0.134 -.0541925 .007264
dg_cie_10_~c | .0404616 .018528 2.18 0.029 .0041474 .0767758
sud_sever~10 | -.0985535 .0423189 -2.33 0.020 -.181497 -.01561
macrozone | .2418558 .024309 9.95 0.000 .1942109 .2895006
policonsumo | .0771658 .0487001 1.58 0.113 -.0182847 .1726163
n_off_vio | .4120495 .0365221 11.28 0.000 .3404675 .4836314
n_off_acq | 1.114617 .0336735 33.10 0.000 1.048618 1.180616
n_off_sud | .3350629 .0358019 9.36 0.000 .2648925 .4052334
clas | .1137162 .028639 3.97 0.000 .0575848 .1698475
motivodeeg~( | -.2712167 .0235993 -11.49 0.000 -.3174704 -.224963
motivodeeg~( | .070912 .0089868 7.89 0.000 .0532983 .0885258
_cons | -6.545356 .2612413 -25.05 0.000 -7.057379 -6.033332
------------------------------------------------------------------------------
***********
family Cox tvc 3
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_3
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20467.815
Iteration 2: log likelihood = -18709.54
Iteration 3: log likelihood = -18078.671
Iteration 4: log likelihood = -18075.827
Iteration 5: log likelihood = -18075.823
Iteration 6: log likelihood = -18075.823
Survival model Number of obs = 59,755
Log likelihood = -18075.823
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0253549 .0257843 -0.98 0.325 -.0758912 .0251813
edad_al_in~1 | .0149346 .0048727 3.06 0.002 .0053842 .024485
edad_ini_c~s | -.0259697 .0046846 -5.54 0.000 -.0351513 -.0167881
sex_enc | -.5061737 .0441552 -11.46 0.000 -.5927163 -.419631
esc_rec | .2481186 .026463 9.38 0.000 .1962522 .299985
sus_prin_mod | .2182386 .0181578 12.02 0.000 .18265 .2538273
fr_sus_prin | .0362553 .0165867 2.19 0.029 .0037459 .0687647
comp_biosoc | .2208494 .030185 7.32 0.000 .1616878 .2800109
ten_viv | -.0232959 .0156757 -1.49 0.137 -.0540197 .0074279
dg_cie_10_~c | .0398249 .0185304 2.15 0.032 .003506 .0761439
sud_sever~10 | -.1017041 .0423323 -2.40 0.016 -.1846738 -.0187343
macrozone | .2425313 .0243141 9.97 0.000 .1948766 .290186
policonsumo | .0744654 .0486615 1.53 0.126 -.0209095 .1698402
n_off_vio | .4115074 .0365233 11.27 0.000 .339923 .4830918
n_off_acq | 1.11462 .0336652 33.11 0.000 1.048637 1.180603
n_off_sud | .3375699 .0358054 9.43 0.000 .2673927 .4077472
clas | .1127797 .0286381 3.94 0.000 .0566501 .1689093
motivodeeg~( | -.2974882 .0259687 -11.46 0.000 -.3483859 -.2465906
motivodeeg~( | .0925428 .0120461 7.68 0.000 .0689328 .1161528
motivodeeg~( | .0217847 .0066895 3.26 0.001 .0086735 .034896
_cons | -6.581578 .2625657 -25.07 0.000 -7.096198 -6.066959
------------------------------------------------------------------------------
***********
family Cox tvc 4
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_4
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20467.241
Iteration 2: log likelihood = -18705.031
Iteration 3: log likelihood = -18079.15
Iteration 4: log likelihood = -18075.612
Iteration 5: log likelihood = -18075.6
Iteration 6: log likelihood = -18075.6
Survival model Number of obs = 59,755
Log likelihood = -18075.6
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0235826 .0260433 -0.91 0.365 -.0746265 .0274614
edad_al_in~1 | .0147069 .0048938 3.01 0.003 .0051153 .0242985
edad_ini_c~s | -.0259282 .0046841 -5.54 0.000 -.0351088 -.0167475
sex_enc | -.5064588 .0441572 -11.47 0.000 -.5930054 -.4199122
esc_rec | .2482672 .0264667 9.38 0.000 .1963934 .300141
sus_prin_mod | .2180646 .0181613 12.01 0.000 .1824691 .25366
fr_sus_prin | .0363112 .0165856 2.19 0.029 .0038041 .0688184
comp_biosoc | .2206645 .030186 7.31 0.000 .1615011 .2798279
ten_viv | -.0232964 .0156769 -1.49 0.137 -.0540226 .0074298
dg_cie_10_~c | .0398433 .0185304 2.15 0.032 .0035244 .0761623
sud_sever~10 | -.101847 .042334 -2.41 0.016 -.1848201 -.0188739
macrozone | .2424027 .0243145 9.97 0.000 .1947471 .2900584
policonsumo | .0748274 .0486718 1.54 0.124 -.0205676 .1702223
n_off_vio | .4119481 .0365289 11.28 0.000 .3403527 .4835435
n_off_acq | 1.115051 .0336741 33.11 0.000 1.049051 1.181051
n_off_sud | .3380121 .0358116 9.44 0.000 .2678227 .4082015
clas | .1127038 .0286374 3.94 0.000 .0565755 .168832
motivodeeg~( | -.2942638 .0269376 -10.92 0.000 -.3470606 -.241467
motivodeeg~( | .089277 .0136757 6.53 0.000 .0624732 .1160808
motivodeeg~( | .0208376 .0077381 2.69 0.007 .0056713 .0360039
motivodeeg~( | .0014435 .0071669 0.20 0.840 -.0126034 .0154903
_cons | -6.574735 .2628672 -25.01 0.000 -7.089946 -6.059525
------------------------------------------------------------------------------
***********
family Cox tvc 5
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_5
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20467.243
Iteration 2: log likelihood = -18703.882
Iteration 3: log likelihood = -18079.377
Iteration 4: log likelihood = -18075.709
Iteration 5: log likelihood = -18075.695
Iteration 6: log likelihood = -18075.695
Survival model Number of obs = 59,755
Log likelihood = -18075.695
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0245235 .02623 -0.93 0.350 -.0759332 .0268863
edad_al_in~1 | .014835 .0049131 3.02 0.003 .0052056 .0244645
edad_ini_c~s | -.0259397 .0046844 -5.54 0.000 -.035121 -.0167584
sex_enc | -.5063726 .0441573 -11.47 0.000 -.5929194 -.4198259
esc_rec | .2482384 .0264661 9.38 0.000 .1963658 .300111
sus_prin_mod | .2181423 .0181622 12.01 0.000 .182545 .2537396
fr_sus_prin | .0363229 .0165868 2.19 0.029 .0038134 .0688324
comp_biosoc | .2206666 .030188 7.31 0.000 .1614992 .279834
ten_viv | -.0232836 .0156765 -1.49 0.137 -.0540091 .0074419
dg_cie_10_~c | .0398204 .0185305 2.15 0.032 .0035013 .0761396
sud_sever~10 | -.1018858 .0423362 -2.41 0.016 -.1848633 -.0189083
macrozone | .2424042 .024316 9.97 0.000 .1947457 .2900626
policonsumo | .074717 .0486698 1.54 0.125 -.020674 .170108
n_off_vio | .4118762 .0365316 11.27 0.000 .3402755 .4834769
n_off_acq | 1.11489 .0336741 33.11 0.000 1.04889 1.18089
n_off_sud | .3379176 .0358132 9.44 0.000 .2677251 .4081102
clas | .1127325 .0286376 3.94 0.000 .0566037 .1688612
motivodeeg~( | -.2961179 .0277026 -10.69 0.000 -.350414 -.2418217
motivodeeg~( | .0907403 .0147651 6.15 0.000 .0618013 .1196793
motivodeeg~( | .0219405 .0077575 2.83 0.005 .006736 .037145
motivodeeg~( | .0059782 .0079157 0.76 0.450 -.0095363 .0214927
motivodeeg~( | .0012469 .0076795 0.16 0.871 -.0138046 .0162985
_cons | -6.578614 .2632298 -24.99 0.000 -7.094535 -6.062693
------------------------------------------------------------------------------
***********
family Cox tvc 6
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_6
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20466.967
Iteration 2: log likelihood = -18705.391
Iteration 3: log likelihood = -18078.202
Iteration 4: log likelihood = -18074.185
Iteration 5: log likelihood = -18074.162
Iteration 6: log likelihood = -18074.162
Survival model Number of obs = 59,755
Log likelihood = -18074.162
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0277658 .0264481 -1.05 0.294 -.079603 .0240715
edad_al_in~1 | .0151266 .0049345 3.07 0.002 .0054551 .0247981
edad_ini_c~s | -.0259008 .004685 -5.53 0.000 -.0350832 -.0167183
sex_enc | -.5066378 .0441566 -11.47 0.000 -.5931831 -.4200925
esc_rec | .2486398 .0264721 9.39 0.000 .1967554 .3005242
sus_prin_mod | .2184342 .0181657 12.02 0.000 .1828301 .2540384
fr_sus_prin | .0362963 .0165875 2.19 0.029 .0037855 .0688071
comp_biosoc | .2207212 .0301839 7.31 0.000 .1615619 .2798804
ten_viv | -.0234423 .0156763 -1.50 0.135 -.0541672 .0072827
dg_cie_10_~c | .0400624 .0185313 2.16 0.031 .0037417 .0763831
sud_sever~10 | -.1012925 .0423361 -2.39 0.017 -.1842697 -.0183154
macrozone | .2425192 .0243195 9.97 0.000 .1948539 .2901845
policonsumo | .0735254 .0486629 1.51 0.131 -.0218522 .168903
n_off_vio | .4113522 .0365321 11.26 0.000 .3397506 .4829537
n_off_acq | 1.114934 .0336724 33.11 0.000 1.048938 1.180931
n_off_sud | .3375462 .0358105 9.43 0.000 .2673589 .4077335
clas | .112708 .0286406 3.94 0.000 .0565735 .1688426
motivodeeg~( | -.302455 .0284002 -10.65 0.000 -.3581184 -.2467917
motivodeeg~( | .0972553 .0157633 6.17 0.000 .0663597 .1281509
motivodeeg~( | .0251361 .0078291 3.21 0.001 .0097914 .0404808
motivodeeg~( | .0123522 .0078726 1.57 0.117 -.0030778 .0277821
motivodeeg~( | .0034626 .0081939 0.42 0.673 -.0125972 .0195225
motivodeeg~( | .0121514 .0081748 1.49 0.137 -.0038708 .0281737
_cons | -6.58897 .2636875 -24.99 0.000 -7.105788 -6.072152
------------------------------------------------------------------------------
***********
family Cox tvc 7
***********
note; a delayed entry model is being fitted
variables created for model 1, component 18: _cmp_1_18_1 to _cmp_1_18_7
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20466.895
Iteration 2: log likelihood = -18706.238
Iteration 3: log likelihood = -18078.396
Iteration 4: log likelihood = -18074.203
Iteration 5: log likelihood = -18074.178
Iteration 6: log likelihood = -18074.178
Survival model Number of obs = 59,755
Log likelihood = -18074.178
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0280413 .0265598 -1.06 0.291 -.0800977 .024015
edad_al_in~1 | .0151507 .0049413 3.07 0.002 .005466 .0248353
edad_ini_c~s | -.0259001 .0046852 -5.53 0.000 -.0350829 -.0167173
sex_enc | -.5066151 .0441569 -11.47 0.000 -.5931609 -.4200692
esc_rec | .2486391 .0264726 9.39 0.000 .1967538 .3005245
sus_prin_mod | .2184772 .0181666 12.03 0.000 .1828713 .254083
fr_sus_prin | .036309 .0165877 2.19 0.029 .0037977 .0688202
comp_biosoc | .2207143 .0301844 7.31 0.000 .1615539 .2798747
ten_viv | -.0234803 .0156762 -1.50 0.134 -.0542051 .0072445
dg_cie_10_~c | .0400799 .0185315 2.16 0.031 .0037589 .0764009
sud_sever~10 | -.10119 .042336 -2.39 0.017 -.184167 -.0182131
macrozone | .2425635 .02432 9.97 0.000 .1948971 .2902298
policonsumo | .0734771 .0486642 1.51 0.131 -.021903 .1688573
n_off_vio | .4112923 .0365323 11.26 0.000 .3396903 .4828943
n_off_acq | 1.114856 .0336721 33.11 0.000 1.04886 1.180852
n_off_sud | .3375061 .0358105 9.42 0.000 .2673187 .4076935
clas | .1126863 .0286409 3.93 0.000 .0565512 .1688214
motivodeeg~( | -.3030218 .0287804 -10.53 0.000 -.3594304 -.2466132
motivodeeg~( | .0978117 .0163626 5.98 0.000 .0657416 .1298817
motivodeeg~( | .0247419 .0078269 3.16 0.002 .0094016 .0400823
motivodeeg~( | .0144652 .0079015 1.83 0.067 -.0010214 .0299519
motivodeeg~( | .0026042 .0079372 0.33 0.743 -.0129524 .0181608
motivodeeg~( | .0100515 .0083687 1.20 0.230 -.006351 .0264539
motivodeeg~( | .0080955 .0086512 0.94 0.349 -.0088606 .0250516
_cons | -6.589793 .2638208 -24.98 0.000 -7.106872 -6.072713
------------------------------------------------------------------------------
.
. // Gompertz
. di in yellow "{bf: ***********}"
***********
. di in yellow "{bf: family Gomp}"
family Gomp
. di in yellow "{bf: ***********}"
***********
. set seed 2125
. qui cap noi stmerlin $covs_2 , dist(gompertz)
note; a delayed entry model is being fitted
Fitting full model:
Iteration 0: log likelihood = -23175439
Iteration 1: log likelihood = -25328.979
Iteration 2: log likelihood = -22602.3
Iteration 3: log likelihood = -18181.413
Iteration 4: log likelihood = -17934.309
Iteration 5: log likelihood = -17922.851
Iteration 6: log likelihood = -17922.814
Iteration 7: log likelihood = -17922.814
Survival model Number of obs = 59,755
Log likelihood = -17922.814
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1330271 .0216996 6.13 0.000 .0904967 .1755576
edad_al_in~1 | .150792 .0095375 15.81 0.000 .1320988 .1694852
edad_ini_c~s | -.023687 .0046503 -5.09 0.000 -.0328014 -.0145727
sex_enc | -.5003202 .044156 -11.33 0.000 -.5868644 -.4137759
esc_rec | .2404719 .0263237 9.14 0.000 .1888784 .2920654
sus_prin_mod | .2267276 .0182075 12.45 0.000 .1910415 .2624137
fr_sus_prin | .0303638 .0166098 1.83 0.068 -.0021909 .0629185
comp_biosoc | .2295395 .0302129 7.60 0.000 .1703234 .2887556
ten_viv | -.0240712 .0156446 -1.54 0.124 -.0547341 .0065918
dg_cie_10_~c | .0403695 .0186965 2.16 0.031 .0037251 .0770139
sud_sever~10 | -.1069822 .0422713 -2.53 0.011 -.1898324 -.0241321
macrozone | .2354406 .024337 9.67 0.000 .187741 .2831403
policonsumo | .1169943 .048632 2.41 0.016 .0216773 .2123114
n_off_vio | .3731639 .0365234 10.22 0.000 .3015795 .4447484
n_off_acq | 1.058839 .0337356 31.39 0.000 .9927182 1.124959
n_off_sud | .3049674 .0357986 8.52 0.000 .2348035 .3751313
clas | .1184432 .0286173 4.14 0.000 .0623542 .1745322
_cons | -4.44128 .2149435 -20.66 0.000 -4.862561 -4.019998
gamma | -.1875864 .0092909 -20.19 0.000 -.2057963 -.1693764
------------------------------------------------------------------------------
. //qui cap noi merlin (_time $covs if _trans == 1, family(gompertz, fail(_status)))
. estimates store m2_1_gom
.
. // Weibull
. di in yellow "{bf: ***********}"
***********
. di in yellow "{bf: family Weibull}"
family Weibull
. di in yellow "{bf: ***********}"
***********
. set seed 2125
. qui cap noi stmerlin $covs_2 , dist(weibull)
note; a delayed entry model is being fitted
Fitting full model:
Iteration 0: log likelihood = -233401.56
Iteration 1: log likelihood = -20488.609 (not concave)
Iteration 2: log likelihood = -18605.226
Iteration 3: log likelihood = -18162.559
Iteration 4: log likelihood = -18143.04 (not concave)
Iteration 5: log likelihood = -18142.613
Iteration 6: log likelihood = -18137.651
Iteration 7: log likelihood = -18131.722
Iteration 8: log likelihood = -18121.609
Iteration 9: log likelihood = -18114.671
Iteration 10: log likelihood = -18106.354
Iteration 11: log likelihood = -18104.292
Iteration 12: log likelihood = -18101.836
Iteration 13: log likelihood = -18100.554
Iteration 14: log likelihood = -18098.394
Iteration 15: log likelihood = -18097.427 (not concave)
Iteration 16: log likelihood = -18097.416
Iteration 17: log likelihood = -18097.171
Iteration 18: log likelihood = -18096.61
Iteration 19: log likelihood = -18096.34
Iteration 20: log likelihood = -18096.167
Iteration 21: log likelihood = -18095.933
Iteration 22: log likelihood = -18095.916
Iteration 23: log likelihood = -18095.914
Iteration 24: log likelihood = -18095.913
Iteration 25: log likelihood = -18095.913
Iteration 26: log likelihood = -18095.913 (not concave)
Iteration 27: log likelihood = -18095.913 (not concave)
Iteration 28: log likelihood = -18095.913 (not concave)
Iteration 29: log likelihood = -18095.913 (not concave)
Iteration 30: log likelihood = -18095.913 (not concave)
Iteration 31: log likelihood = -18095.913 (not concave)
Iteration 32: log likelihood = -18095.913 (not concave)
Iteration 33: log likelihood = -18095.913 (not concave)
Iteration 34: log likelihood = -18095.913 (not concave)
Iteration 35: log likelihood = -18095.913 (not concave)
Iteration 36: log likelihood = -18095.913 (not concave)
Iteration 37: log likelihood = -18095.913 (not concave)
Iteration 38: log likelihood = -18095.913 (not concave)
Iteration 39: log likelihood = -18095.913 (not concave)
Iteration 40: log likelihood = -18095.913 (not concave)
Iteration 41: log likelihood = -18095.913 (not concave)
Iteration 42: log likelihood = -18095.913 (not concave)
Iteration 43: log likelihood = -18095.913 (not concave)
Iteration 44: log likelihood = -18095.913 (not concave)
Iteration 45: log likelihood = -18095.913 (not concave)
Iteration 46: log likelihood = -18095.913 (not concave)
Iteration 47: log likelihood = -18095.913 (not concave)
Iteration 48: log likelihood = -18095.913 (not concave)
Iteration 49: log likelihood = -18095.913 (not concave)
Iteration 50: log likelihood = -18095.913 (not concave)
Iteration 51: log likelihood = -18095.913 (not concave)
Iteration 52: log likelihood = -18095.913 (not concave)
Iteration 53: log likelihood = -18095.913 (not concave)
Iteration 54: log likelihood = -18095.913 (not concave)
Iteration 55: log likelihood = -18095.913 (not concave)
Iteration 56: log likelihood = -18095.913 (not concave)
Iteration 57: log likelihood = -18095.913 (not concave)
Iteration 58: log likelihood = -18095.913 (not concave)
Iteration 59: log likelihood = -18095.913 (not concave)
Iteration 60: log likelihood = -18095.913 (not concave)
Iteration 61: log likelihood = -18095.913 (not concave)
Iteration 62: log likelihood = -18095.913 (not concave)
Iteration 63: log likelihood = -18095.913 (not concave)
Iteration 64: log likelihood = -18095.913 (not concave)
Iteration 65: log likelihood = -18095.913 (not concave)
Iteration 66: log likelihood = -18095.913 (not concave)
Iteration 67: log likelihood = -18095.913 (not concave)
Iteration 68: log likelihood = -18095.913 (not concave)
Iteration 69: log likelihood = -18095.913 (not concave)
Iteration 70: log likelihood = -18095.913 (not concave)
Iteration 71: log likelihood = -18095.913 (not concave)
Iteration 72: log likelihood = -18095.913 (not concave)
Iteration 73: log likelihood = -18095.913 (not concave)
Iteration 74: log likelihood = -18095.913 (not concave)
Iteration 75: log likelihood = -18095.913 (not concave)
Iteration 76: log likelihood = -18095.913 (not concave)
Iteration 77: log likelihood = -18095.913 (not concave)
Iteration 78: log likelihood = -18095.913 (not concave)
Iteration 79: log likelihood = -18095.913 (not concave)
Iteration 80: log likelihood = -18095.913 (not concave)
Iteration 81: log likelihood = -18095.913 (not concave)
Iteration 82: log likelihood = -18095.913 (not concave)
Iteration 83: log likelihood = -18095.913 (not concave)
Iteration 84: log likelihood = -18095.913 (not concave)
Iteration 85: log likelihood = -18095.913 (not concave)
Iteration 86: log likelihood = -18095.913 (not concave)
Iteration 87: log likelihood = -18095.913 (not concave)
Iteration 88: log likelihood = -18095.913 (not concave)
Iteration 89: log likelihood = -18095.913 (not concave)
Iteration 90: log likelihood = -18095.913 (not concave)
Iteration 91: log likelihood = -18095.913 (not concave)
Iteration 92: log likelihood = -18095.913 (not concave)
Iteration 93: log likelihood = -18095.913 (not concave)
Iteration 94: log likelihood = -18095.913 (not concave)
Iteration 95: log likelihood = -18095.913 (not concave)
Iteration 96: log likelihood = -18095.913 (not concave)
Iteration 97: log likelihood = -18095.913 (not concave)
Iteration 98: log likelihood = -18095.913 (not concave)
Iteration 99: log likelihood = -18095.913 (not concave)
Iteration 100: log likelihood = -18095.913 (not concave)
Iteration 101: log likelihood = -18095.913 (not concave)
Iteration 102: log likelihood = -18095.913 (not concave)
Iteration 103: log likelihood = -18095.913 (not concave)
Iteration 104: log likelihood = -18095.913 (not concave)
Iteration 105: log likelihood = -18095.913 (not concave)
Iteration 106: log likelihood = -18095.913 (not concave)
Iteration 107: log likelihood = -18095.913 (not concave)
Iteration 108: log likelihood = -18095.913 (not concave)
Iteration 109: log likelihood = -18095.913 (not concave)
Iteration 110: log likelihood = -18095.913 (not concave)
Iteration 111: log likelihood = -18095.913 (not concave)
Iteration 112: log likelihood = -18095.913 (not concave)
Iteration 113: log likelihood = -18095.913 (not concave)
Iteration 114: log likelihood = -18095.913 (not concave)
Iteration 115: log likelihood = -18095.913 (not concave)
Iteration 116: log likelihood = -18095.913 (not concave)
Iteration 117: log likelihood = -18095.913 (not concave)
Iteration 118: log likelihood = -18095.913 (not concave)
Iteration 119: log likelihood = -18095.913 (not concave)
Iteration 120: log likelihood = -18095.913 (not concave)
Iteration 121: log likelihood = -18095.913 (not concave)
Iteration 122: log likelihood = -18095.913 (not concave)
Iteration 123: log likelihood = -18095.913 (not concave)
Iteration 124: log likelihood = -18095.913 (not concave)
Iteration 125: log likelihood = -18095.913 (not concave)
Iteration 126: log likelihood = -18095.913 (not concave)
Iteration 127: log likelihood = -18095.913 (not concave)
Iteration 128: log likelihood = -18095.913 (not concave)
Iteration 129: log likelihood = -18095.913 (not concave)
Iteration 130: log likelihood = -18095.913 (not concave)
Iteration 131: log likelihood = -18095.913 (not concave)
Iteration 132: log likelihood = -18095.913 (not concave)
Iteration 133: log likelihood = -18095.913 (not concave)
Iteration 134: log likelihood = -18095.913 (not concave)
Iteration 135: log likelihood = -18095.913 (not concave)
Iteration 136: log likelihood = -18095.913 (not concave)
Iteration 137: log likelihood = -18095.913 (not concave)
Iteration 138: log likelihood = -18095.913 (not concave)
Iteration 139: log likelihood = -18095.913 (not concave)
Iteration 140: log likelihood = -18095.913 (not concave)
Iteration 141: log likelihood = -18095.913 (not concave)
Iteration 142: log likelihood = -18095.913 (not concave)
Iteration 143: log likelihood = -18095.913 (not concave)
Iteration 144: log likelihood = -18095.913 (not concave)
Iteration 145: log likelihood = -18095.913 (not concave)
Iteration 146: log likelihood = -18095.913 (not concave)
Iteration 147: log likelihood = -18095.913 (not concave)
Iteration 148: log likelihood = -18095.913 (not concave)
Iteration 149: log likelihood = -18095.913 (not concave)
Iteration 150: log likelihood = -18095.913 (not concave)
Iteration 151: log likelihood = -18095.913 (not concave)
Iteration 152: log likelihood = -18095.913 (not concave)
Iteration 153: log likelihood = -18095.913 (not concave)
Iteration 154: log likelihood = -18095.913 (not concave)
Iteration 155: log likelihood = -18095.913 (not concave)
Iteration 156: log likelihood = -18095.913 (not concave)
Iteration 157: log likelihood = -18095.913 (not concave)
Iteration 158: log likelihood = -18095.913 (not concave)
Iteration 159: log likelihood = -18095.913 (not concave)
Iteration 160: log likelihood = -18095.913 (not concave)
Iteration 161: log likelihood = -18095.913 (not concave)
Iteration 162: log likelihood = -18095.913 (not concave)
Iteration 163: log likelihood = -18095.913 (not concave)
Iteration 164: log likelihood = -18095.913 (not concave)
Iteration 165: log likelihood = -18095.913 (not concave)
Iteration 166: log likelihood = -18095.913 (not concave)
Iteration 167: log likelihood = -18095.913 (not concave)
Iteration 168: log likelihood = -18095.913 (not concave)
Iteration 169: log likelihood = -18095.913 (not concave)
Iteration 170: log likelihood = -18095.913 (not concave)
Iteration 171: log likelihood = -18095.913 (not concave)
Iteration 172: log likelihood = -18095.913 (not concave)
Iteration 173: log likelihood = -18095.913 (not concave)
Iteration 174: log likelihood = -18095.913 (not concave)
Iteration 175: log likelihood = -18095.913 (not concave)
Iteration 176: log likelihood = -18095.913 (not concave)
Iteration 177: log likelihood = -18095.913 (not concave)
Iteration 178: log likelihood = -18095.913 (not concave)
Iteration 179: log likelihood = -18095.913 (not concave)
Iteration 180: log likelihood = -18095.913 (not concave)
Iteration 181: log likelihood = -18095.913 (not concave)
Iteration 182: log likelihood = -18095.913 (not concave)
Iteration 183: log likelihood = -18095.913 (not concave)
Iteration 184: log likelihood = -18095.913 (not concave)
Iteration 185: log likelihood = -18095.913 (not concave)
Iteration 186: log likelihood = -18095.913 (not concave)
Iteration 187: log likelihood = -18095.913 (not concave)
Iteration 188: log likelihood = -18095.913 (not concave)
Iteration 189: log likelihood = -18095.913 (not concave)
Iteration 190: log likelihood = -18095.913 (not concave)
Iteration 191: log likelihood = -18095.913 (not concave)
Iteration 192: log likelihood = -18095.913 (not concave)
Iteration 193: log likelihood = -18095.913 (not concave)
Iteration 194: log likelihood = -18095.913 (not concave)
Iteration 195: log likelihood = -18095.913 (not concave)
Iteration 196: log likelihood = -18095.913 (not concave)
Iteration 197: log likelihood = -18095.913 (not concave)
Iteration 198: log likelihood = -18095.913 (not concave)
Iteration 199: log likelihood = -18095.913 (not concave)
Iteration 200: log likelihood = -18095.913 (not concave)
Iteration 201: log likelihood = -18095.913 (not concave)
Iteration 202: log likelihood = -18095.913 (not concave)
Iteration 203: log likelihood = -18095.913 (not concave)
Iteration 204: log likelihood = -18095.913 (not concave)
Iteration 205: log likelihood = -18095.913 (not concave)
Iteration 206: log likelihood = -18095.913 (not concave)
Iteration 207: log likelihood = -18095.913 (not concave)
Iteration 208: log likelihood = -18095.913 (not concave)
Iteration 209: log likelihood = -18095.913 (not concave)
Iteration 210: log likelihood = -18095.913 (not concave)
Iteration 211: log likelihood = -18095.913 (not concave)
Iteration 212: log likelihood = -18095.913 (not concave)
Iteration 213: log likelihood = -18095.913 (not concave)
Iteration 214: log likelihood = -18095.913 (not concave)
Iteration 215: log likelihood = -18095.913 (not concave)
Iteration 216: log likelihood = -18095.913 (not concave)
Iteration 217: log likelihood = -18095.913 (not concave)
Iteration 218: log likelihood = -18095.913 (not concave)
Iteration 219: log likelihood = -18095.913 (not concave)
Iteration 220: log likelihood = -18095.913 (not concave)
Iteration 221: log likelihood = -18095.913 (not concave)
Iteration 222: log likelihood = -18095.913 (not concave)
Iteration 223: log likelihood = -18095.913 (not concave)
Iteration 224: log likelihood = -18095.913 (not concave)
Iteration 225: log likelihood = -18095.913 (not concave)
Iteration 226: log likelihood = -18095.913 (not concave)
Iteration 227: log likelihood = -18095.913 (not concave)
Iteration 228: log likelihood = -18095.913 (not concave)
Iteration 229: log likelihood = -18095.913 (not concave)
Iteration 230: log likelihood = -18095.913 (not concave)
Iteration 231: log likelihood = -18095.913 (not concave)
Iteration 232: log likelihood = -18095.913 (not concave)
Iteration 233: log likelihood = -18095.913 (not concave)
Iteration 234: log likelihood = -18095.913 (not concave)
Iteration 235: log likelihood = -18095.913 (not concave)
Iteration 236: log likelihood = -18095.913 (not concave)
Iteration 237: log likelihood = -18095.913 (not concave)
Iteration 238: log likelihood = -18095.913 (not concave)
Iteration 239: log likelihood = -18095.913 (not concave)
Iteration 240: log likelihood = -18095.913 (not concave)
Iteration 241: log likelihood = -18095.913 (not concave)
Iteration 242: log likelihood = -18095.913 (not concave)
Iteration 243: log likelihood = -18095.913 (not concave)
Iteration 244: log likelihood = -18095.913 (not concave)
Iteration 245: log likelihood = -18095.913 (not concave)
Iteration 246: log likelihood = -18095.913 (not concave)
Iteration 247: log likelihood = -18095.913 (not concave)
Iteration 248: log likelihood = -18095.913 (not concave)
Iteration 249: log likelihood = -18095.913 (not concave)
Iteration 250: log likelihood = -18095.913 (not concave)
Iteration 251: log likelihood = -18095.913 (not concave)
Iteration 252: log likelihood = -18095.913 (not concave)
Iteration 253: log likelihood = -18095.913 (not concave)
Iteration 254: log likelihood = -18095.913 (not concave)
Iteration 255: log likelihood = -18095.913 (not concave)
Iteration 256: log likelihood = -18095.913 (not concave)
Iteration 257: log likelihood = -18095.913 (not concave)
Iteration 258: log likelihood = -18095.913 (not concave)
Iteration 259: log likelihood = -18095.913 (not concave)
Iteration 260: log likelihood = -18095.913 (not concave)
Iteration 261: log likelihood = -18095.913 (not concave)
Iteration 262: log likelihood = -18095.913 (not concave)
Iteration 263: log likelihood = -18095.913 (not concave)
Iteration 264: log likelihood = -18095.913 (not concave)
Iteration 265: log likelihood = -18095.913 (not concave)
Iteration 266: log likelihood = -18095.913 (not concave)
Iteration 267: log likelihood = -18095.913 (not concave)
Iteration 268: log likelihood = -18095.913 (not concave)
Iteration 269: log likelihood = -18095.913 (not concave)
Iteration 270: log likelihood = -18095.913 (not concave)
Iteration 271: log likelihood = -18095.913 (not concave)
Iteration 272: log likelihood = -18095.913 (not concave)
Iteration 273: log likelihood = -18095.913 (not concave)
Iteration 274: log likelihood = -18095.913 (not concave)
Iteration 275: log likelihood = -18095.913 (not concave)
Iteration 276: log likelihood = -18095.913 (not concave)
Iteration 277: log likelihood = -18095.913 (not concave)
Iteration 278: log likelihood = -18095.913 (not concave)
Iteration 279: log likelihood = -18095.913 (not concave)
Iteration 280: log likelihood = -18095.913 (not concave)
Iteration 281: log likelihood = -18095.913 (not concave)
Iteration 282: log likelihood = -18095.913 (not concave)
Iteration 283: log likelihood = -18095.913 (not concave)
Iteration 284: log likelihood = -18095.913 (not concave)
Iteration 285: log likelihood = -18095.913 (not concave)
Iteration 286: log likelihood = -18095.913 (not concave)
Iteration 287: log likelihood = -18095.913 (not concave)
Iteration 288: log likelihood = -18095.913 (not concave)
Iteration 289: log likelihood = -18095.913 (not concave)
Iteration 290: log likelihood = -18095.913 (not concave)
Iteration 291: log likelihood = -18095.913 (not concave)
Iteration 292: log likelihood = -18095.913 (not concave)
Iteration 293: log likelihood = -18095.913 (not concave)
Iteration 294: log likelihood = -18095.913 (not concave)
Iteration 295: log likelihood = -18095.913 (not concave)
Iteration 296: log likelihood = -18095.913 (not concave)
Iteration 297: log likelihood = -18095.913 (not concave)
Iteration 298: log likelihood = -18095.913 (not concave)
Iteration 299: log likelihood = -18095.913 (not concave)
Iteration 300: log likelihood = -18095.913 (not concave)
convergence not achieved
Survival model Number of obs = 59,755
Log likelihood = -18095.913
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1317154 .021417 6.15 0.000 .0897388 .173692
edad_al_in~1 | -.008856 .002061 -4.30 0.000 -.0128955 -.0048165
edad_ini_c~s | -.0255288 .0046395 -5.50 0.000 -.0346221 -.0164355
sex_enc | -.5067476 .0441787 -11.47 0.000 -.5933362 -.420159
esc_rec | .2307585 .0263562 8.76 0.000 .1791013 .2824157
sus_prin_mod | .2169924 .0180854 12.00 0.000 .1815456 .2524392
fr_sus_prin | .0340983 .0166141 2.05 0.040 .0015353 .0666613
comp_biosoc | .2192777 .0301563 7.27 0.000 .1601724 .278383
ten_viv | -.0185395 .0156644 -1.18 0.237 -.0492412 .0121621
dg_cie_10_~c | .0397738 .0185807 2.14 0.032 .0033564 .0761912
sud_sever~10 | -.107603 .0422691 -2.55 0.011 -.1904488 -.0247571
macrozone | .2385112 .0243282 9.80 0.000 .1908288 .2861936
policonsumo | .082049 .0485853 1.69 0.091 -.0131765 .1772744
n_off_vio | .417209 .0364886 11.43 0.000 .3456926 .4887253
n_off_acq | 1.119769 .0336202 33.31 0.000 1.053874 1.185663
n_off_sud | .3490227 .0357809 9.75 0.000 .2788934 .419152
clas | .1076992 .0286249 3.76 0.000 .0515955 .1638029
_cons | 14.38984 . . . . .
log(gamma) | -16.6463 .2137268 -77.89 0.000 -17.0652 -16.2274
------------------------------------------------------------------------------
. //qui cap noi merlin (_time $covs if _trans == 1, family(gompertz, fail(_status)))
. estimates store m2_1_wei
.
. // Log logistic
. di in yellow "{bf: ***********}"
***********
. di in yellow "{bf: family Logl}"
family Logl
. di in yellow "{bf: ***********}"
***********
. set seed 2125
. qui cap noi stmerlin $covs_2 , dist(loglogistic)
note; a delayed entry model is being fitted
Fitting full model:
Iteration 0: log likelihood = -20064.873 (not concave)
Iteration 1: log likelihood = -20033.19 (not concave)
Iteration 2: log likelihood = -18734.102 (not concave)
Iteration 3: log likelihood = -18504.793 (not concave)
Iteration 4: log likelihood = -18410.588 (not concave)
Iteration 5: log likelihood = -18366.73 (not concave)
Iteration 6: log likelihood = -18346.993 (not concave)
Iteration 7: log likelihood = -18326.771 (not concave)
Iteration 8: log likelihood = -18317.54 (not concave)
Iteration 9: log likelihood = -18305.826 (not concave)
Iteration 10: log likelihood = -18294.415 (not concave)
Iteration 11: log likelihood = -18288.176
Iteration 12: log likelihood = -18199.328
Iteration 13: log likelihood = -18140.751
Iteration 14: log likelihood = -18136.237
Iteration 15: log likelihood = -18135.484
Iteration 16: log likelihood = -18135.48
Iteration 17: log likelihood = -18135.48
Survival model Number of obs = 59,755
Log likelihood = -18135.48
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.0974722 .0165511 -5.89 0.000 -.1299118 -.0650327
edad_al_in~1 | .0357222 .0016083 22.21 0.000 .0325701 .0388743
edad_ini_c~s | .0161467 .0031915 5.06 0.000 .0098915 .0224019
sex_enc | .3716758 .0364986 10.18 0.000 .3001399 .4432116
esc_rec | -.1647651 .0212071 -7.77 0.000 -.2063302 -.1231999
sus_prin_mod | -.1484725 .015564 -9.54 0.000 -.1789774 -.1179676
fr_sus_prin | -.0342671 .0121928 -2.81 0.005 -.0581645 -.0103697
comp_biosoc | -.1564546 .0237645 -6.58 0.000 -.2030322 -.109877
ten_viv | .0153271 .0122025 1.26 0.209 -.0085894 .0392436
dg_cie_10_~c | -.0217444 .0142456 -1.53 0.127 -.0496652 .0061765
sud_sever~10 | .057354 .0306366 1.87 0.061 -.0026926 .1174007
macrozone | -.1750096 .0220358 -7.94 0.000 -.2181989 -.1318202
policonsumo | -.0286542 .034173 -0.84 0.402 -.095632 .0383235
n_off_vio | -.4167489 .0404012 -10.32 0.000 -.4959337 -.3375641
n_off_acq | -.9952754 .0719441 -13.83 0.000 -1.136283 -.8542677
n_off_sud | -.3448578 .0374426 -9.21 0.000 -.4182439 -.2714716
clas | -.0611817 .021535 -2.84 0.004 -.1033896 -.0189738
_cons | 4.24267 .1617643 26.23 0.000 3.925618 4.559722
dap:1 | -.81126 .0382933 -21.19 0.000 -.8863135 -.7362065
------------------------------------------------------------------------------
. //qui cap noi merlin (_time $covs if _trans == 1, family(loglogistic, fail(_status)))
. estimates store m2_1_logl
.
. // Log normal
. di in yellow "{bf: ***********}"
***********
. di in yellow "{bf: family Logn}"
family Logn
. di in yellow "{bf: ***********}"
***********
. set seed 2125
. qui cap noi stmerlin $covs_2 , dist(lognormal)
note; a delayed entry model is being fitted
Fitting full model:
Iteration 0: log likelihood = -33053.961 (not concave)
Iteration 1: log likelihood = -21475.955 (not concave)
Iteration 2: log likelihood = -19430.343 (not concave)
Iteration 3: log likelihood = -18719.789
Iteration 4: log likelihood = -18414.254 (backed up)
Iteration 5: log likelihood = -18246.139
Iteration 6: log likelihood = -18171.858
Iteration 7: log likelihood = -18145.434
Iteration 8: log likelihood = -18115.354
Iteration 9: log likelihood = -18115.248
Iteration 10: log likelihood = -18097.462
Iteration 11: log likelihood = -18094.582
Iteration 12: log likelihood = -18091.98
Iteration 13: log likelihood = -18091.415
Iteration 14: log likelihood = -18090.586
Iteration 15: log likelihood = -18090.468
Iteration 16: log likelihood = -18090.445
Iteration 17: log likelihood = -18090.444
Survival model Number of obs = 59,755
Log likelihood = -18090.444
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.2863444 .0707408 -4.05 0.000 -.4249939 -.147695
edad_al_in~1 | .052058 .0067099 7.76 0.000 .0389068 .0652092
edad_ini_c~s | .0424814 .0114114 3.72 0.000 .0201155 .0648472
sex_enc | 1.139331 .2308911 4.93 0.000 .6867926 1.591869
esc_rec | -.5055517 .11013 -4.59 0.000 -.7214025 -.2897009
sus_prin_mod | -.477813 .099348 -4.81 0.000 -.6725315 -.2830945
fr_sus_prin | -.1212626 .0441213 -2.75 0.006 -.2077389 -.0347864
comp_biosoc | -.4615491 .108843 -4.24 0.000 -.6748774 -.2482208
ten_viv | .0512756 .0397362 1.29 0.197 -.0266058 .1291571
dg_cie_10_~c | -.0733538 .0464226 -1.58 0.114 -.1643405 .0176329
sud_sever~10 | .1532806 .0959881 1.60 0.110 -.0348527 .3414139
macrozone | -.5577172 .1226181 -4.55 0.000 -.7980442 -.3173901
policonsumo | -.12277 .1059545 -1.16 0.247 -.330437 .084897
n_off_vio | -1.339563 .2857668 -4.69 0.000 -1.899656 -.7794704
n_off_acq | -3.41275 .7199551 -4.74 0.000 -4.823837 -2.001664
n_off_sud | -1.147916 .2525861 -4.54 0.000 -1.642976 -.6528565
clas | -.2041612 .07502 -2.72 0.007 -.3511978 -.0571246
_cons | 6.063686 .6085038 9.96 0.000 4.871041 7.256332
dap:1 | .4687489 .1237405 3.79 0.000 .2262219 .7112759
------------------------------------------------------------------------------
. //qui cap noi merlin (_time $covs if _trans == 1, family(lognormal, fail(_status)))
. estimates store m2_1_logn
.
. // Generalised gamma
. di in yellow "{bf: ***********}"
***********
. di in yellow "{bf: family Ggam}"
family Ggam
. di in yellow "{bf: ***********}"
***********
. set seed 2125
. qui cap noi stmerlin $covs_2 , dist(ggamma)
note; a delayed entry model is being fitted
Fitting full model:
Iteration 0: log likelihood = -21405.054 (not concave)
Iteration 1: log likelihood = -19350.697 (not concave)
Iteration 2: log likelihood = -18762.239 (not concave)
Iteration 3: log likelihood = -18543.466
Iteration 4: log likelihood = -18299.243 (not concave)
Iteration 5: log likelihood = -18282.705 (not concave)
Iteration 6: log likelihood = -18189.688 (not concave)
Iteration 7: log likelihood = -18146.109 (not concave)
Iteration 8: log likelihood = -18109.262 (not concave)
Iteration 9: log likelihood = -18096.961
Iteration 10: log likelihood = -18084.54
Iteration 11: log likelihood = -18080.785
Iteration 12: log likelihood = -18076.61 (not concave)
Iteration 13: log likelihood = -18075.569 (not concave)
Iteration 14: log likelihood = -18075.389 (not concave)
Iteration 15: log likelihood = -18075.244 (not concave)
Iteration 16: log likelihood = -18075.123 (not concave)
Iteration 17: log likelihood = -18075.029
Iteration 18: log likelihood = -18074.115 (not concave)
Iteration 19: log likelihood = -18072.707 (not concave)
Iteration 20: log likelihood = -18072.542 (not concave)
Iteration 21: log likelihood = -18072.387 (not concave)
Iteration 22: log likelihood = -18072.311 (not concave)
Iteration 23: log likelihood = -18072.258 (not concave)
Iteration 24: log likelihood = -18072.219 (not concave)
Iteration 25: log likelihood = -18072.187 (not concave)
Iteration 26: log likelihood = -18072.165 (not concave)
Iteration 27: log likelihood = -18072.138 (not concave)
Iteration 28: log likelihood = -18072.114
Iteration 29: log likelihood = -18072.025 (not concave)
Iteration 30: log likelihood = -18071.937 (not concave)
Iteration 31: log likelihood = -18071.899
Iteration 32: log likelihood = -18071.872 (not concave)
Iteration 33: log likelihood = -18071.848 (not concave)
Iteration 34: log likelihood = -18071.815 (not concave)
Iteration 35: log likelihood = -18071.795 (not concave)
Iteration 36: log likelihood = -18071.775 (not concave)
Iteration 37: log likelihood = -18071.753 (not concave)
Iteration 38: log likelihood = -18071.721 (not concave)
Iteration 39: log likelihood = -18071.717
Iteration 40: log likelihood = -18071.684 (not concave)
Iteration 41: log likelihood = -18071.578 (not concave)
Iteration 42: log likelihood = -18071.531 (not concave)
Iteration 43: log likelihood = -18071.482 (not concave)
Iteration 44: log likelihood = -18071.459 (not concave)
Iteration 45: log likelihood = -18071.451 (not concave)
Iteration 46: log likelihood = -18071.448 (not concave)
Iteration 47: log likelihood = -18071.236 (not concave)
Iteration 48: log likelihood = -18071.212 (not concave)
Iteration 49: log likelihood = -18071.195 (not concave)
Iteration 50: log likelihood = -18071.18 (not concave)
Iteration 51: log likelihood = -18071.169 (not concave)
Iteration 52: log likelihood = -18071.13 (not concave)
Iteration 53: log likelihood = -18071.122 (not concave)
Iteration 54: log likelihood = -18071.117 (not concave)
Iteration 55: log likelihood = -18071.112 (not concave)
Iteration 56: log likelihood = -18071.108 (not concave)
Iteration 57: log likelihood = -18071.092 (not concave)
Iteration 58: log likelihood = -18071.084 (not concave)
Iteration 59: log likelihood = -18071.065 (not concave)
Iteration 60: log likelihood = -18071.033 (not concave)
Iteration 61: log likelihood = -18071.013 (not concave)
Iteration 62: log likelihood = -18070.997 (not concave)
Iteration 63: log likelihood = -18070.973 (not concave)
Iteration 64: log likelihood = -18070.963 (not concave)
Iteration 65: log likelihood = -18070.926 (not concave)
Iteration 66: log likelihood = -18070.907 (not concave)
Iteration 67: log likelihood = -18070.879 (not concave)
Iteration 68: log likelihood = -18070.872 (not concave)
Iteration 69: log likelihood = -18070.859 (not concave)
Iteration 70: log likelihood = -18070.846 (not concave)
Iteration 71: log likelihood = -18070.842 (not concave)
Iteration 72: log likelihood = -18070.841 (not concave)
Iteration 73: log likelihood = -18070.838 (not concave)
Iteration 74: log likelihood = -18070.837 (not concave)
Iteration 75: log likelihood = -18070.808 (not concave)
Iteration 76: log likelihood = -18070.8 (not concave)
Iteration 77: log likelihood = -18070.793 (not concave)
Iteration 78: log likelihood = -18070.788 (not concave)
Iteration 79: log likelihood = -18070.784 (not concave)
Iteration 80: log likelihood = -18070.781 (not concave)
Iteration 81: log likelihood = -18070.779 (not concave)
Iteration 82: log likelihood = -18070.778 (not concave)
Iteration 83: log likelihood = -18070.761 (not concave)
Iteration 84: log likelihood = -18070.745 (not concave)
Iteration 85: log likelihood = -18070.739 (not concave)
Iteration 86: log likelihood = -18070.736 (not concave)
Iteration 87: log likelihood = -18070.733 (not concave)
Iteration 88: log likelihood = -18070.717 (not concave)
Iteration 89: log likelihood = -18070.712 (not concave)
Iteration 90: log likelihood = -18070.708 (not concave)
Iteration 91: log likelihood = -18070.703 (not concave)
Iteration 92: log likelihood = -18070.702 (not concave)
Iteration 93: log likelihood = -18070.699 (not concave)
Iteration 94: log likelihood = -18070.696 (not concave)
Iteration 95: log likelihood = -18070.694 (not concave)
Iteration 96: log likelihood = -18070.692 (not concave)
Iteration 97: log likelihood = -18070.677 (not concave)
Iteration 98: log likelihood = -18070.674 (not concave)
Iteration 99: log likelihood = -18070.664 (not concave)
Iteration 100: log likelihood = -18070.656 (not concave)
Iteration 101: log likelihood = -18070.65 (not concave)
Iteration 102: log likelihood = -18070.646 (not concave)
Iteration 103: log likelihood = -18070.644 (not concave)
Iteration 104: log likelihood = -18070.638 (not concave)
Iteration 105: log likelihood = -18070.631 (not concave)
Iteration 106: log likelihood = -18070.628 (not concave)
Iteration 107: log likelihood = -18070.627 (not concave)
Iteration 108: log likelihood = -18070.615 (not concave)
Iteration 109: log likelihood = -18070.609 (not concave)
Iteration 110: log likelihood = -18070.609 (not concave)
Iteration 111: log likelihood = -18070.599 (not concave)
Iteration 112: log likelihood = -18070.59 (not concave)
Iteration 113: log likelihood = -18070.585 (not concave)
Iteration 114: log likelihood = -18070.584 (not concave)
Iteration 115: log likelihood = -18070.583 (not concave)
Iteration 116: log likelihood = -18070.557 (not concave)
Iteration 117: log likelihood = -18070.55 (not concave)
Iteration 118: log likelihood = -18070.542 (not concave)
Iteration 119: log likelihood = -18070.539 (not concave)
Iteration 120: log likelihood = -18070.538 (not concave)
Iteration 121: log likelihood = -18070.538 (not concave)
Iteration 122: log likelihood = -18070.537 (not concave)
Iteration 123: log likelihood = -18070.537 (not concave)
Iteration 124: log likelihood = -18070.535 (not concave)
Iteration 125: log likelihood = -18070.534 (not concave)
Iteration 126: log likelihood = -18070.532 (not concave)
Iteration 127: log likelihood = -18070.529 (not concave)
Iteration 128: log likelihood = -18070.528 (not concave)
Iteration 129: log likelihood = -18070.526 (not concave)
Iteration 130: log likelihood = -18070.523 (not concave)
Iteration 131: log likelihood = -18070.52 (not concave)
Iteration 132: log likelihood = -18070.516 (not concave)
Iteration 133: log likelihood = -18070.513 (not concave)
Iteration 134: log likelihood = -18070.513 (not concave)
Iteration 135: log likelihood = -18070.511 (not concave)
Iteration 136: log likelihood = -18070.51 (not concave)
Iteration 137: log likelihood = -18070.508 (not concave)
Iteration 138: log likelihood = -18070.506 (not concave)
Iteration 139: log likelihood = -18070.505 (not concave)
Iteration 140: log likelihood = -18070.498 (not concave)
Iteration 141: log likelihood = -18070.495 (not concave)
Iteration 142: log likelihood = -18070.492 (not concave)
Iteration 143: log likelihood = -18070.489 (not concave)
Iteration 144: log likelihood = -18070.479 (not concave)
Iteration 145: log likelihood = -18070.472 (not concave)
Iteration 146: log likelihood = -18070.469 (not concave)
Iteration 147: log likelihood = -18070.467 (not concave)
Iteration 148: log likelihood = -18070.467 (not concave)
Iteration 149: log likelihood = -18070.467 (not concave)
Iteration 150: log likelihood = -18070.442 (not concave)
Iteration 151: log likelihood = -18070.435 (not concave)
Iteration 152: log likelihood = -18070.433 (not concave)
Iteration 153: log likelihood = -18070.433 (not concave)
Iteration 154: log likelihood = -18070.433 (not concave)
Iteration 155: log likelihood = -18070.433 (not concave)
Iteration 156: log likelihood = -18070.41 (not concave)
Iteration 157: log likelihood = -18070.408 (not concave)
Iteration 158: log likelihood = -18070.4 (not concave)
Iteration 159: log likelihood = -18070.4 (not concave)
Iteration 160: log likelihood = -18070.39 (not concave)
Iteration 161: log likelihood = -18070.384 (not concave)
Iteration 162: log likelihood = -18070.38 (not concave)
Iteration 163: log likelihood = -18070.378 (not concave)
Iteration 164: log likelihood = -18070.364 (not concave)
Iteration 165: log likelihood = -18070.361 (not concave)
Iteration 166: log likelihood = -18070.359 (not concave)
Iteration 167: log likelihood = -18070.359 (not concave)
Iteration 168: log likelihood = -18070.359 (not concave)
Iteration 169: log likelihood = -18070.359 (not concave)
Iteration 170: log likelihood = -18070.358 (not concave)
Iteration 171: log likelihood = -18070.358 (not concave)
Iteration 172: log likelihood = -18070.358 (not concave)
Iteration 173: log likelihood = -18070.358 (not concave)
Iteration 174: log likelihood = -18070.358 (not concave)
Iteration 175: log likelihood = -18070.358 (not concave)
Iteration 176: log likelihood = -18070.358 (not concave)
Iteration 177: log likelihood = -18070.358 (not concave)
Iteration 178: log likelihood = -18070.358 (not concave)
Iteration 179: log likelihood = -18070.358 (not concave)
Iteration 180: log likelihood = -18070.358 (not concave)
Iteration 181: log likelihood = -18070.358 (not concave)
Iteration 182: log likelihood = -18070.358 (not concave)
Iteration 183: log likelihood = -18070.358 (not concave)
Iteration 184: log likelihood = -18070.358 (not concave)
Iteration 185: log likelihood = -18070.358 (not concave)
Iteration 186: log likelihood = -18070.358 (not concave)
Iteration 187: log likelihood = -18070.358 (not concave)
Iteration 188: log likelihood = -18070.358 (not concave)
Iteration 189: log likelihood = -18070.358 (not concave)
Iteration 190: log likelihood = -18070.358 (not concave)
Iteration 191: log likelihood = -18070.358 (not concave)
Iteration 192: log likelihood = -18070.358 (not concave)
Iteration 193: log likelihood = -18070.358 (not concave)
Iteration 194: log likelihood = -18070.358 (not concave)
Iteration 195: log likelihood = -18070.358 (not concave)
Iteration 196: log likelihood = -18070.358 (not concave)
Iteration 197: log likelihood = -18070.358 (not concave)
Iteration 198: log likelihood = -18070.358 (not concave)
Iteration 199: log likelihood = -18070.358 (not concave)
Iteration 200: log likelihood = -18070.358 (not concave)
Iteration 201: log likelihood = -18070.358 (not concave)
Iteration 202: log likelihood = -18070.358 (not concave)
Iteration 203: log likelihood = -18070.358 (not concave)
Iteration 204: log likelihood = -18070.358 (not concave)
Iteration 205: log likelihood = -18070.358 (not concave)
Iteration 206: log likelihood = -18070.358 (not concave)
Iteration 207: log likelihood = -18070.358 (not concave)
Iteration 208: log likelihood = -18070.358 (not concave)
Iteration 209: log likelihood = -18070.358 (not concave)
Iteration 210: log likelihood = -18070.358 (not concave)
Iteration 211: log likelihood = -18070.358 (not concave)
Iteration 212: log likelihood = -18070.358 (not concave)
Iteration 213: log likelihood = -18070.358 (not concave)
Iteration 214: log likelihood = -18070.358 (not concave)
Iteration 215: log likelihood = -18070.358 (not concave)
Iteration 216: log likelihood = -18070.358 (not concave)
Iteration 217: log likelihood = -18070.358 (not concave)
Iteration 218: log likelihood = -18070.358 (not concave)
Iteration 219: log likelihood = -18070.358 (not concave)
Iteration 220: log likelihood = -18070.358 (not concave)
Iteration 221: log likelihood = -18070.358 (not concave)
Iteration 222: log likelihood = -18070.358 (not concave)
Iteration 223: log likelihood = -18070.358 (not concave)
Iteration 224: log likelihood = -18070.358 (not concave)
Iteration 225: log likelihood = -18070.358 (not concave)
Iteration 226: log likelihood = -18070.358 (not concave)
Iteration 227: log likelihood = -18070.358 (not concave)
Iteration 228: log likelihood = -18070.358 (not concave)
Iteration 229: log likelihood = -18070.358 (not concave)
Iteration 230: log likelihood = -18070.358 (not concave)
Iteration 231: log likelihood = -18070.358 (not concave)
Iteration 232: log likelihood = -18070.358 (not concave)
Iteration 233: log likelihood = -18070.358 (not concave)
Iteration 234: log likelihood = -18070.358 (not concave)
Iteration 235: log likelihood = -18070.358 (not concave)
Iteration 236: log likelihood = -18070.358 (not concave)
Iteration 237: log likelihood = -18070.358 (not concave)
Iteration 238: log likelihood = -18070.358 (not concave)
Iteration 239: log likelihood = -18070.358 (not concave)
Iteration 240: log likelihood = -18070.358 (not concave)
Iteration 241: log likelihood = -18070.358 (not concave)
Iteration 242: log likelihood = -18070.358 (not concave)
Iteration 243: log likelihood = -18070.358 (not concave)
Iteration 244: log likelihood = -18070.358 (not concave)
Iteration 245: log likelihood = -18070.358 (not concave)
Iteration 246: log likelihood = -18070.358 (not concave)
Iteration 247: log likelihood = -18070.358 (not concave)
Iteration 248: log likelihood = -18070.358 (not concave)
Iteration 249: log likelihood = -18070.358 (not concave)
Iteration 250: log likelihood = -18070.358 (not concave)
Iteration 251: log likelihood = -18070.358 (not concave)
Iteration 252: log likelihood = -18070.358 (not concave)
Iteration 253: log likelihood = -18070.358 (not concave)
Iteration 254: log likelihood = -18070.358 (not concave)
Iteration 255: log likelihood = -18070.358 (not concave)
Iteration 256: log likelihood = -18070.358 (not concave)
Iteration 257: log likelihood = -18070.358 (not concave)
Iteration 258: log likelihood = -18070.358 (not concave)
Iteration 259: log likelihood = -18070.358 (not concave)
Iteration 260: log likelihood = -18070.358 (not concave)
Iteration 261: log likelihood = -18070.358 (not concave)
Iteration 262: log likelihood = -18070.358 (not concave)
Iteration 263: log likelihood = -18070.358 (not concave)
Iteration 264: log likelihood = -18070.358 (not concave)
Iteration 265: log likelihood = -18070.358 (not concave)
Iteration 266: log likelihood = -18070.358 (not concave)
Iteration 267: log likelihood = -18070.358 (not concave)
Iteration 268: log likelihood = -18070.358 (not concave)
Iteration 269: log likelihood = -18070.358 (not concave)
Iteration 270: log likelihood = -18070.358 (not concave)
Iteration 271: log likelihood = -18070.358 (not concave)
Iteration 272: log likelihood = -18070.358 (not concave)
Iteration 273: log likelihood = -18070.358 (not concave)
Iteration 274: log likelihood = -18070.358 (not concave)
Iteration 275: log likelihood = -18070.358 (not concave)
Iteration 276: log likelihood = -18070.358 (not concave)
Iteration 277: log likelihood = -18070.358 (not concave)
Iteration 278: log likelihood = -18070.358 (not concave)
Iteration 279: log likelihood = -18070.358 (not concave)
Iteration 280: log likelihood = -18070.358 (not concave)
Iteration 281: log likelihood = -18070.358 (not concave)
Iteration 282: log likelihood = -18070.358 (not concave)
Iteration 283: log likelihood = -18070.358 (not concave)
Iteration 284: log likelihood = -18070.358 (not concave)
Iteration 285: log likelihood = -18070.358 (not concave)
Iteration 286: log likelihood = -18070.358 (not concave)
Iteration 287: log likelihood = -18070.358 (not concave)
Iteration 288: log likelihood = -18070.358 (not concave)
Iteration 289: log likelihood = -18070.358 (not concave)
Iteration 290: log likelihood = -18070.358 (not concave)
Iteration 291: log likelihood = -18070.358 (not concave)
Iteration 292: log likelihood = -18070.358 (not concave)
Iteration 293: log likelihood = -18070.358 (not concave)
Iteration 294: log likelihood = -18070.358 (not concave)
Iteration 295: log likelihood = -18070.358 (not concave)
Iteration 296: log likelihood = -18070.358 (not concave)
Iteration 297: log likelihood = -18070.358 (not concave)
Iteration 298: log likelihood = -18070.358 (not concave)
Iteration 299: log likelihood = -18070.358 (not concave)
Iteration 300: log likelihood = -18070.358 (not concave)
convergence not achieved
Survival model Number of obs = 59,755
Log likelihood = -18070.358
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | -.4661748 .0275377 -16.93 0.000 -.5201477 -.4122019
edad_al_in~1 | .0649589 .0037795 17.19 0.000 .0575512 .0723666
edad_ini_c~s | .0757888 .0071795 10.56 0.000 .0617172 .0898604
sex_enc | 1.840372 .0410558 44.83 0.000 1.759904 1.92084
esc_rec | -.8171341 . . . . .
sus_prin_mod | -.7780551 . . . . .
fr_sus_prin | -.1625322 .027628 -5.88 0.000 -.2166821 -.1083823
comp_biosoc | -.7608284 . . . . .
ten_viv | .0812349 .0153993 5.28 0.000 .0510529 .111417
dg_cie_10_~c | -.1255617 .0354895 -3.54 0.000 -.1951199 -.0560035
sud_sever~10 | .3114935 .0428679 7.27 0.000 .227474 .3955129
macrozone | -.8913888 .0915375 -9.74 0.000 -1.070799 -.7119786
policonsumo | -.2233535 .0941391 -2.37 0.018 -.4078627 -.0388443
n_off_vio | -1.910195 .21573 -8.85 0.000 -2.333018 -1.487372
n_off_acq | -4.699182 . . . . .
n_off_sud | -1.60925 . . . . .
clas | -.3429518 .0222005 -15.45 0.000 -.3864639 -.2994397
_cons | 7.01491 .0832339 84.28 0.000 6.851775 7.178046
log(sigma) | .8961913 .015385 58.25 0.000 .8660372 .9263455
kappa | .2764748 . . . . .
------------------------------------------------------------------------------
. //qui cap noi merlin (_time $covs if _trans == 1, family(ggamma, fail(_status)))
. estimates store m2_1_ggam
.
. // Royston Parmar models
. forvalues j=1/10 {
2. di in yellow "{bf: ***********}"
3. di in yellow "{bf: family RP`j'}"
4. di in yellow "{bf: ***********}"
5. set seed 2125
6. qui cap noi stmerlin $covs_2, dist(rp) df(`j')
7. //qui cap noi merlin (_time $covs if _trans == 1, family(rp, df(`j') fail(_status)))
. estimates store m2_1_rp`j'
8. *estimates save "${pathdata2}parmodels.ster", append
. }
***********
family RP1
***********
note; a delayed entry model is being fitted
variables created: _rcs1_1 to _rcs1_1
Fitting full model:
Iteration 0: log likelihood = -36826.906 (not concave)
Iteration 1: log likelihood = -20926.082
Iteration 2: log likelihood = -20071.583 (not concave)
Iteration 3: log likelihood = -18560.105 (not concave)
Iteration 4: log likelihood = -18258.367
Iteration 5: log likelihood = -18134.596 (not concave)
Iteration 6: log likelihood = -18131.637
Iteration 7: log likelihood = -18125.268
Iteration 8: log likelihood = -18118.35
Iteration 9: log likelihood = -18110.519
Iteration 10: log likelihood = -18107.682
Iteration 11: log likelihood = -18104.471
Iteration 12: log likelihood = -18102.79
Iteration 13: log likelihood = -18101.569
Iteration 14: log likelihood = -18100.502
Iteration 15: log likelihood = -18099.187
Iteration 16: log likelihood = -18098.932
Iteration 17: log likelihood = -18098.541
Iteration 18: log likelihood = -18098.088
Iteration 19: log likelihood = -18097.878
Iteration 20: log likelihood = -18097.554
Iteration 21: log likelihood = -18097.453
Iteration 22: log likelihood = -18097.291
Iteration 23: log likelihood = -18097.105
Iteration 24: log likelihood = -18096.986
Iteration 25: log likelihood = -18096.878
Iteration 26: log likelihood = -18096.798
Iteration 27: log likelihood = -18096.654
Iteration 28: log likelihood = -18096.6
Iteration 29: log likelihood = -18096.532
Iteration 30: log likelihood = -18096.462
Iteration 31: log likelihood = -18096.46
Iteration 32: log likelihood = -18096.391 (not concave)
Iteration 33: log likelihood = -18096.391
Iteration 34: log likelihood = -18096.371
Iteration 35: log likelihood = -18096.338
Iteration 36: log likelihood = -18096.302
Iteration 37: log likelihood = -18096.259
Iteration 38: log likelihood = -18096.215
Iteration 39: log likelihood = -18096.204
Iteration 40: log likelihood = -18096.188
Iteration 41: log likelihood = -18096.159
Iteration 42: log likelihood = -18096.139
Iteration 43: log likelihood = -18096.131
Iteration 44: log likelihood = -18096.106
Iteration 45: log likelihood = -18096.101
Iteration 46: log likelihood = -18096.085
Iteration 47: log likelihood = -18096.063
Iteration 48: log likelihood = -18096.049
Iteration 49: log likelihood = -18096.045
Iteration 50: log likelihood = -18096.037
Iteration 51: log likelihood = -18096.027
Iteration 52: log likelihood = -18096.022
Iteration 53: log likelihood = -18096.016
Iteration 54: log likelihood = -18096.01
Iteration 55: log likelihood = -18096.005
Iteration 56: log likelihood = -18095.999
Iteration 57: log likelihood = -18095.999
Iteration 58: log likelihood = -18095.993 (not concave)
Iteration 59: log likelihood = -18095.993
Iteration 60: log likelihood = -18095.991 (backed up)
Iteration 61: log likelihood = -18095.989
Iteration 62: log likelihood = -18095.981
Iteration 63: log likelihood = -18095.978
Iteration 64: log likelihood = -18095.975
Iteration 65: log likelihood = -18095.97
Iteration 66: log likelihood = -18095.969
Iteration 67: log likelihood = -18095.965
Iteration 68: log likelihood = -18095.964
Iteration 69: log likelihood = -18095.962
Iteration 70: log likelihood = -18095.959
Iteration 71: log likelihood = -18095.958
Iteration 72: log likelihood = -18095.955
Iteration 73: log likelihood = -18095.953 (not concave)
Iteration 74: log likelihood = -18095.953
Iteration 75: log likelihood = -18095.953
Iteration 76: log likelihood = -18095.951 (not concave)
Iteration 77: log likelihood = -18095.95 (not concave)
Iteration 78: log likelihood = -18095.95
Iteration 79: log likelihood = -18095.95
Iteration 80: log likelihood = -18095.947
Iteration 81: log likelihood = -18095.947
Iteration 82: log likelihood = -18095.946
Iteration 83: log likelihood = -18095.945
Iteration 84: log likelihood = -18095.944
Iteration 85: log likelihood = -18095.942
Iteration 86: log likelihood = -18095.941
Iteration 87: log likelihood = -18095.94
Iteration 88: log likelihood = -18095.94
Iteration 89: log likelihood = -18095.938 (not concave)
Iteration 90: log likelihood = -18095.938
Iteration 91: log likelihood = -18095.938
Iteration 92: log likelihood = -18095.937 (not concave)
Iteration 93: log likelihood = -18095.937 (backed up)
Iteration 94: log likelihood = -18095.936
Iteration 95: log likelihood = -18095.935 (not concave)
Iteration 96: log likelihood = -18095.935
Iteration 97: log likelihood = -18095.935 (backed up)
Iteration 98: log likelihood = -18095.934
Iteration 99: log likelihood = -18095.933
Iteration 100: log likelihood = -18095.932
Iteration 101: log likelihood = -18095.932
Iteration 102: log likelihood = -18095.931
Iteration 103: log likelihood = -18095.931
Iteration 104: log likelihood = -18095.93
Iteration 105: log likelihood = -18095.93
Iteration 106: log likelihood = -18095.929 (not concave)
Iteration 107: log likelihood = -18095.929
Iteration 108: log likelihood = -18095.928
Iteration 109: log likelihood = -18095.928
Iteration 110: log likelihood = -18095.928
Iteration 111: log likelihood = -18095.927
Iteration 112: log likelihood = -18095.927
Iteration 113: log likelihood = -18095.926
Iteration 114: log likelihood = -18095.926
Iteration 115: log likelihood = -18095.926 (not concave)
Iteration 116: log likelihood = -18095.926
Iteration 117: log likelihood = -18095.925
Iteration 118: log likelihood = -18095.925
Iteration 119: log likelihood = -18095.925
Iteration 120: log likelihood = -18095.925
Iteration 121: log likelihood = -18095.924 (not concave)
Iteration 122: log likelihood = -18095.924 (backed up)
Iteration 123: log likelihood = -18095.924
Iteration 124: log likelihood = -18095.924
Iteration 125: log likelihood = -18095.923
Iteration 126: log likelihood = -18095.923
Iteration 127: log likelihood = -18095.923
Iteration 128: log likelihood = -18095.923
Iteration 129: log likelihood = -18095.922
Iteration 130: log likelihood = -18095.922
Iteration 131: log likelihood = -18095.922
Iteration 132: log likelihood = -18095.922
Iteration 133: log likelihood = -18095.921
Iteration 134: log likelihood = -18095.921
Iteration 135: log likelihood = -18095.921
Iteration 136: log likelihood = -18095.921
Iteration 137: log likelihood = -18095.92 (not concave)
Iteration 138: log likelihood = -18095.92
Iteration 139: log likelihood = -18095.92
Iteration 140: log likelihood = -18095.92
Iteration 141: log likelihood = -18095.92
Iteration 142: log likelihood = -18095.92
Iteration 143: log likelihood = -18095.92
Iteration 144: log likelihood = -18095.919
Iteration 145: log likelihood = -18095.919
Iteration 146: log likelihood = -18095.919
Iteration 147: log likelihood = -18095.919
Iteration 148: log likelihood = -18095.919
Iteration 149: log likelihood = -18095.919
Iteration 150: log likelihood = -18095.919
Iteration 151: log likelihood = -18095.918
Iteration 152: log likelihood = -18095.918
Iteration 153: log likelihood = -18095.918
Iteration 154: log likelihood = -18095.918
Iteration 155: log likelihood = -18095.918
Iteration 156: log likelihood = -18095.918
Iteration 157: log likelihood = -18095.918
Iteration 158: log likelihood = -18095.918
Iteration 159: log likelihood = -18095.918
Iteration 160: log likelihood = -18095.917
Iteration 161: log likelihood = -18095.917
Iteration 162: log likelihood = -18095.917
Iteration 163: log likelihood = -18095.917
Iteration 164: log likelihood = -18095.917
Iteration 165: log likelihood = -18095.917
Iteration 166: log likelihood = -18095.917 (not concave)
Iteration 167: log likelihood = -18095.917 (backed up)
Iteration 168: log likelihood = -18095.917
Iteration 169: log likelihood = -18095.917
Iteration 170: log likelihood = -18095.917
Iteration 171: log likelihood = -18095.916
Iteration 172: log likelihood = -18095.916
Iteration 173: log likelihood = -18095.916
Iteration 174: log likelihood = -18095.916
Iteration 175: log likelihood = -18095.916
Iteration 176: log likelihood = -18095.916
Iteration 177: log likelihood = -18095.916
Iteration 178: log likelihood = -18095.916
Iteration 179: log likelihood = -18095.916
Iteration 180: log likelihood = -18095.916
Iteration 181: log likelihood = -18095.916
Iteration 182: log likelihood = -18095.916
Iteration 183: log likelihood = -18095.916
Iteration 184: log likelihood = -18095.916
Iteration 185: log likelihood = -18095.916
Iteration 186: log likelihood = -18095.916
Iteration 187: log likelihood = -18095.916 (not concave)
Iteration 188: log likelihood = -18095.916
Iteration 189: log likelihood = -18095.916
Iteration 190: log likelihood = -18095.916
Iteration 191: log likelihood = -18095.915
Iteration 192: log likelihood = -18095.915
Iteration 193: log likelihood = -18095.915
Iteration 194: log likelihood = -18095.915
Iteration 195: log likelihood = -18095.915
Iteration 196: log likelihood = -18095.915
Iteration 197: log likelihood = -18095.915
Iteration 198: log likelihood = -18095.915
Iteration 199: log likelihood = -18095.915
Iteration 200: log likelihood = -18095.915
Iteration 201: log likelihood = -18095.915
Iteration 202: log likelihood = -18095.915
Iteration 203: log likelihood = -18095.915
Iteration 204: log likelihood = -18095.915
Iteration 205: log likelihood = -18095.915
Iteration 206: log likelihood = -18095.915 (not concave)
Iteration 207: log likelihood = -18095.915
Iteration 208: log likelihood = -18095.915
Iteration 209: log likelihood = -18095.915
Iteration 210: log likelihood = -18095.915
Iteration 211: log likelihood = -18095.915
Iteration 212: log likelihood = -18095.915
Iteration 213: log likelihood = -18095.915 (not concave)
Iteration 214: log likelihood = -18095.915 (backed up)
Iteration 215: log likelihood = -18095.915
Iteration 216: log likelihood = -18095.915
Iteration 217: log likelihood = -18095.915
Iteration 218: log likelihood = -18095.915
Iteration 219: log likelihood = -18095.915
Iteration 220: log likelihood = -18095.915
Iteration 221: log likelihood = -18095.915 (not concave)
Iteration 222: log likelihood = -18095.915 (backed up)
Iteration 223: log likelihood = -18095.914
Iteration 224: log likelihood = -18095.914
Iteration 225: log likelihood = -18095.914
Iteration 226: log likelihood = -18095.914
Iteration 227: log likelihood = -18095.914 (not concave)
Iteration 228: log likelihood = -18095.914 (backed up)
Iteration 229: log likelihood = -18095.914
Iteration 230: log likelihood = -18095.914
Iteration 231: log likelihood = -18095.914
Iteration 232: log likelihood = -18095.914
Iteration 233: log likelihood = -18095.914
Iteration 234: log likelihood = -18095.914
Iteration 235: log likelihood = -18095.914
Iteration 236: log likelihood = -18095.914
Iteration 237: log likelihood = -18095.914 (not concave)
Iteration 238: log likelihood = -18095.914 (backed up)
Iteration 239: log likelihood = -18095.914 (backed up)
Iteration 240: log likelihood = -18095.914
Iteration 241: log likelihood = -18095.914
Iteration 242: log likelihood = -18095.914 (not concave)
Iteration 243: log likelihood = -18095.914
Iteration 244: log likelihood = -18095.914
Iteration 245: log likelihood = -18095.914
Iteration 246: log likelihood = -18095.914
Iteration 247: log likelihood = -18095.914
Iteration 248: log likelihood = -18095.914
Iteration 249: log likelihood = -18095.914
Iteration 250: log likelihood = -18095.914
Iteration 251: log likelihood = -18095.914
Iteration 252: log likelihood = -18095.914
Iteration 253: log likelihood = -18095.914
Iteration 254: log likelihood = -18095.914 (not concave)
Iteration 255: log likelihood = -18095.914 (not concave)
Iteration 256: log likelihood = -18095.914 (not concave)
Iteration 257: log likelihood = -18095.914 (not concave)
Iteration 258: log likelihood = -18095.914 (not concave)
Iteration 259: log likelihood = -18095.914 (not concave)
Iteration 260: log likelihood = -18095.914 (not concave)
Iteration 261: log likelihood = -18095.914 (not concave)
Iteration 262: log likelihood = -18095.914 (not concave)
Iteration 263: log likelihood = -18095.914 (not concave)
Iteration 264: log likelihood = -18095.914 (not concave)
Iteration 265: log likelihood = -18095.914 (not concave)
Iteration 266: log likelihood = -18095.914 (not concave)
Iteration 267: log likelihood = -18095.914 (not concave)
Iteration 268: log likelihood = -18095.914 (not concave)
Iteration 269: log likelihood = -18095.914 (not concave)
Iteration 270: log likelihood = -18095.914 (not concave)
Iteration 271: log likelihood = -18095.914 (not concave)
Iteration 272: log likelihood = -18095.914 (not concave)
Iteration 273: log likelihood = -18095.914 (not concave)
Iteration 274: log likelihood = -18095.914 (not concave)
Iteration 275: log likelihood = -18095.914 (not concave)
Iteration 276: log likelihood = -18095.914 (not concave)
Iteration 277: log likelihood = -18095.914 (not concave)
Iteration 278: log likelihood = -18095.914 (not concave)
Iteration 279: log likelihood = -18095.914 (not concave)
Iteration 280: log likelihood = -18095.914 (not concave)
Iteration 281: log likelihood = -18095.914 (not concave)
Iteration 282: log likelihood = -18095.914 (not concave)
Iteration 283: log likelihood = -18095.914 (not concave)
Iteration 284: log likelihood = -18095.914 (not concave)
Iteration 285: log likelihood = -18095.914 (not concave)
Iteration 286: log likelihood = -18095.914 (not concave)
Iteration 287: log likelihood = -18095.914 (not concave)
Iteration 288: log likelihood = -18095.914 (not concave)
Iteration 289: log likelihood = -18095.914 (not concave)
Iteration 290: log likelihood = -18095.914 (not concave)
Iteration 291: log likelihood = -18095.914 (not concave)
Iteration 292: log likelihood = -18095.914 (not concave)
Iteration 293: log likelihood = -18095.914 (not concave)
Iteration 294: log likelihood = -18095.914 (not concave)
Iteration 295: log likelihood = -18095.914 (not concave)
Iteration 296: log likelihood = -18095.914 (not concave)
Iteration 297: log likelihood = -18095.914 (not concave)
Iteration 298: log likelihood = -18095.914 (not concave)
Iteration 299: log likelihood = -18095.914 (not concave)
Iteration 300: log likelihood = -18095.914 (not concave)
convergence not achieved
Survival model Number of obs = 59,755
Log likelihood = -18095.914
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1317145 .0214172 6.15 0.000 .0897376 .1736914
edad_al_in~1 | -.0088567 .0020611 -4.30 0.000 -.0128963 -.004817
edad_ini_c~s | -.0255291 .0046396 -5.50 0.000 -.0346225 -.0164357
sex_enc | -.5067487 .0441787 -11.47 0.000 -.5933373 -.4201602
esc_rec | .2307575 .0263563 8.76 0.000 .1791001 .282415
sus_prin_mod | .2169922 .0180855 12.00 0.000 .1815453 .2524391
fr_sus_prin | .0340977 .0166141 2.05 0.040 .0015347 .0666608
comp_biosoc | .2192767 .0301565 7.27 0.000 .160171 .2783823
ten_viv | -.0185402 .0156645 -1.18 0.237 -.049242 .0121617
dg_cie_10_~c | .0397734 .0185806 2.14 0.032 .0033561 .0761907
sud_sever~10 | -.1076052 .0422698 -2.55 0.011 -.1904525 -.0247579
macrozone | .2385104 .0243283 9.80 0.000 .1908278 .286193
policonsumo | .0820477 .0485855 1.69 0.091 -.0131781 .1772734
n_off_vio | .4172096 .0364885 11.43 0.000 .3456935 .4887257
n_off_acq | 1.11977 .03362 33.31 0.000 1.053876 1.185664
n_off_sud | .3490232 .0357808 9.75 0.000 .2788941 .4191522
clas | .1076979 .0286254 3.76 0.000 .0515932 .1638025
_cons | 8.24193 . . . . .
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP2
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_2
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18092.365 (not concave)
Iteration 2: log likelihood = -18090.171 (not concave)
Iteration 3: log likelihood = -18088.781 (not concave)
Iteration 4: log likelihood = -18087.762 (not concave)
Iteration 5: log likelihood = -18086.967 (not concave)
Iteration 6: log likelihood = -18086.209 (not concave)
Iteration 7: log likelihood = -18084.966 (not concave)
Iteration 8: log likelihood = -18083.807 (not concave)
Iteration 9: log likelihood = -18055.047 (not concave)
Iteration 10: log likelihood = -18050.087 (not concave)
Iteration 11: log likelihood = -18048.413 (not concave)
Iteration 12: log likelihood = -18046.74 (not concave)
Iteration 13: log likelihood = -18035.78 (not concave)
Iteration 14: log likelihood = -18024.735 (not concave)
Iteration 15: log likelihood = -18022.121 (not concave)
Iteration 16: log likelihood = -18019.624 (not concave)
Iteration 17: log likelihood = -18017.404 (not concave)
Iteration 18: log likelihood = -18015.1 (not concave)
Iteration 19: log likelihood = -18012.832
Iteration 20: log likelihood = -18000.843 (backed up)
Iteration 21: log likelihood = -17987.168
Iteration 22: log likelihood = -17968.873 (not concave)
Iteration 23: log likelihood = -17968.819
Iteration 24: log likelihood = -17968.799 (not concave)
Iteration 25: log likelihood = -17968.799 (backed up)
Survival model Number of obs = 59,755
Log likelihood = -17968.799
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1295081 .0215156 6.02 0.000 .0873383 .1716779
edad_al_in~1 | .0565853 .0021893 25.85 0.000 .0522942 .0608763
edad_ini_c~s | -.0249519 .0046793 -5.33 0.000 -.0341231 -.0157806
sex_enc | -.5033819 .0441584 -11.40 0.000 -.5899307 -.416833
esc_rec | .2457802 .0263576 9.32 0.000 .1941202 .2974402
sus_prin_mod | .2217259 .0181735 12.20 0.000 .1861065 .2573454
fr_sus_prin | .0341394 .0166014 2.06 0.040 .0016012 .0666775
comp_biosoc | .2227039 .0301768 7.38 0.000 .1635584 .2818493
ten_viv | -.023784 .0156454 -1.52 0.128 -.0544484 .0068804
dg_cie_10_~c | .0400168 .0186269 2.15 0.032 .0035087 .0765249
sud_sever~10 | -.1032074 .042273 -2.44 0.015 -.1860609 -.020354
macrozone | .240902 .0243289 9.90 0.000 .1932181 .2885859
policonsumo | .0855065 .0484852 1.76 0.078 -.0095227 .1805356
n_off_vio | .4002926 .0364487 10.98 0.000 .3288545 .4717307
n_off_acq | 1.095175 .0335546 32.64 0.000 1.029409 1.160941
n_off_sud | .3243698 .0357509 9.07 0.000 .2542994 .3944402
clas | .1132079 .0286261 3.95 0.000 .0571017 .1693141
_cons | 6.441748 6.736474 0.96 0.339 -6.761498 19.64499
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP3
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_3
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18074.139 (not concave)
Iteration 2: log likelihood = -18073.514 (not concave)
Iteration 3: log likelihood = -18064.603 (not concave)
Iteration 4: log likelihood = -18046.931 (not concave)
Iteration 5: log likelihood = -18040.883
Iteration 6: log likelihood = -18015.985
Iteration 7: log likelihood = -17999.501
Iteration 8: log likelihood = -17984.855
Iteration 9: log likelihood = -17969.168
Iteration 10: log likelihood = -17969.102
Iteration 11: log likelihood = -17969.1
Iteration 12: log likelihood = -17969.1 (not concave)
Iteration 13: log likelihood = -17969.1 (backed up)
Iteration 14: log likelihood = -17969.1
Survival model Number of obs = 59,755
Log likelihood = -17969.1
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1293093 .021518 6.01 0.000 .0871348 .1714838
edad_al_in~1 | .0548853 .0032915 16.67 0.000 .0484341 .0613365
edad_ini_c~s | -.0249922 .0046838 -5.34 0.000 -.0341722 -.0158122
sex_enc | -.503372 .044158 -11.40 0.000 -.5899201 -.416824
esc_rec | .246615 .0264639 9.32 0.000 .1947468 .2984832
sus_prin_mod | .2213331 .0181806 12.17 0.000 .1856997 .2569664
fr_sus_prin | .0342162 .016605 2.06 0.039 .001671 .0667614
comp_biosoc | .2223621 .0301815 7.37 0.000 .1632075 .2815166
ten_viv | -.023951 .0156552 -1.53 0.126 -.0546346 .0067327
dg_cie_10_~c | .039956 .018626 2.15 0.032 .0034496 .0764623
sud_sever~10 | -.1031572 .0422808 -2.44 0.015 -.1860261 -.0202883
macrozone | .241375 .0243457 9.91 0.000 .1936584 .2890917
policonsumo | .0838793 .0485609 1.73 0.084 -.0112984 .179057
n_off_vio | .4009949 .0364806 10.99 0.000 .3294943 .4724955
n_off_acq | 1.096215 .033599 32.63 0.000 1.030362 1.162068
n_off_sud | .3248655 .0357493 9.09 0.000 .2547982 .3949329
clas | .1129821 .0286267 3.95 0.000 .0568749 .1690894
_cons | 6.367222 3.065058 2.08 0.038 .359819 12.37463
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP4
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_4
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18074.796 (not concave)
Iteration 2: log likelihood = -18066.769 (not concave)
Iteration 3: log likelihood = -18063.016 (not concave)
Iteration 4: log likelihood = -18057.792 (not concave)
Iteration 5: log likelihood = -18052.926 (not concave)
Iteration 6: log likelihood = -18026.35 (not concave)
Iteration 7: log likelihood = -18023.363 (not concave)
Iteration 8: log likelihood = -18021.04 (not concave)
Iteration 9: log likelihood = -18019.121 (not concave)
Iteration 10: log likelihood = -18017.405 (not concave)
Iteration 11: log likelihood = -18015.826 (not concave)
Iteration 12: log likelihood = -18014.38
Iteration 13: log likelihood = -17999.046
Iteration 14: log likelihood = -17975.684
Iteration 15: log likelihood = -17967.863
Iteration 16: log likelihood = -17967.827 (not concave)
Iteration 17: log likelihood = -17967.827 (not concave)
Iteration 18: log likelihood = -17967.827
Iteration 19: log likelihood = -17967.827
Iteration 20: log likelihood = -17967.827
Survival model Number of obs = 59,755
Log likelihood = -17967.827
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1293869 .0215235 6.01 0.000 .0872016 .1715722
edad_al_in~1 | .0591395 .003936 15.03 0.000 .0514251 .0668539
edad_ini_c~s | -.0249051 .0046781 -5.32 0.000 -.0340741 -.0157361
sex_enc | -.5036204 .0441584 -11.40 0.000 -.5901694 -.4170714
esc_rec | .2469256 .0264774 9.33 0.000 .1950307 .2988204
sus_prin_mod | .2225338 .018195 12.23 0.000 .1868724 .2581953
fr_sus_prin | .0344952 .0166054 2.08 0.038 .0019491 .0670412
comp_biosoc | .223171 .0301834 7.39 0.000 .1640126 .2823293
ten_viv | -.024188 .0156573 -1.54 0.122 -.0548757 .0064996
dg_cie_10_~c | .0403025 .018629 2.16 0.031 .0037903 .0768147
sud_sever~10 | -.1018293 .0422926 -2.41 0.016 -.1847213 -.0189373
macrozone | .2407212 .0243446 9.89 0.000 .1930067 .2884357
policonsumo | .0867405 .0485998 1.78 0.074 -.0085134 .1819944
n_off_vio | .4001823 .0364789 10.97 0.000 .328685 .4716796
n_off_acq | 1.094255 .0336102 32.56 0.000 1.028381 1.16013
n_off_sud | .3221694 .0357883 9.00 0.000 .2520255 .3923133
clas | .1139249 .0286321 3.98 0.000 .057807 .1700428
_cons | 6.425347 9.106904 0.71 0.480 -11.42386 24.27455
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP5
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_5
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18076.917 (not concave)
Iteration 2: log likelihood = -18068.763 (not concave)
Iteration 3: log likelihood = -18063.122 (not concave)
Iteration 4: log likelihood = -18059.484 (not concave)
Iteration 5: log likelihood = -18056.221 (not concave)
Iteration 6: log likelihood = -18053.362
Iteration 7: log likelihood = -18026.92
Iteration 8: log likelihood = -17968.46
Iteration 9: log likelihood = -17963.272
Iteration 10: log likelihood = -17958.553 (not concave)
Iteration 11: log likelihood = -17958.543
Iteration 12: log likelihood = -17958.535
Iteration 13: log likelihood = -17958.535
Iteration 14: log likelihood = -17958.535
Iteration 15: log likelihood = -17958.535 (backed up)
Survival model Number of obs = 59,755
Log likelihood = -17958.535
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .130385 .0215472 6.05 0.000 .0881533 .1726166
edad_al_in~1 | .067934 .004441 15.30 0.000 .0592299 .0766381
edad_ini_c~s | -.0244857 .0046722 -5.24 0.000 -.0336431 -.0153283
sex_enc | -.5046427 .0441558 -11.43 0.000 -.5911865 -.4180989
esc_rec | .2478366 .0265025 9.35 0.000 .1958927 .2997804
sus_prin_mod | .2221272 .0182108 12.20 0.000 .1864346 .2578198
fr_sus_prin | .0340353 .0166042 2.05 0.040 .0014917 .066579
comp_biosoc | .2239229 .0301858 7.42 0.000 .1647598 .283086
ten_viv | -.0246237 .0156644 -1.57 0.116 -.0553253 .006078
dg_cie_10_~c | .0408693 .0186353 2.19 0.028 .0043448 .0773937
sud_sever~10 | -.1008462 .042297 -2.38 0.017 -.1837469 -.0179456
macrozone | .2400879 .0243471 9.86 0.000 .1923684 .2878074
policonsumo | .0909305 .04865 1.87 0.062 -.0044218 .1862827
n_off_vio | .3977987 .0364858 10.90 0.000 .3262878 .4693097
n_off_acq | 1.092395 .0336199 32.49 0.000 1.026501 1.158289
n_off_sud | .3205464 .0357928 8.96 0.000 .2503937 .390699
clas | .1140363 .0286294 3.98 0.000 .0579236 .170149
_cons | 6.093842 4.012833 1.52 0.129 -1.771166 13.95885
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP6
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_6
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18095.59 (not concave)
Iteration 2: log likelihood = -18064.536 (not concave)
Iteration 3: log likelihood = -18057.636 (not concave)
Iteration 4: log likelihood = -18054.561
Iteration 5: log likelihood = -18029.343
Iteration 6: log likelihood = -18005.374
Iteration 7: log likelihood = -17982.951
Iteration 8: log likelihood = -17955.382
Iteration 9: log likelihood = -17955.12
Iteration 10: log likelihood = -17955.093
Iteration 11: log likelihood = -17954.383 (not concave)
Iteration 12: log likelihood = -17954.168
Iteration 13: log likelihood = -17954.119
Iteration 14: log likelihood = -17954.118
Iteration 15: log likelihood = -17954.118
Iteration 16: log likelihood = -17954.118 (backed up)
Survival model Number of obs = 59,755
Log likelihood = -17954.118
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1303047 .0215589 6.04 0.000 .08805 .1725594
edad_al_in~1 | .0737065 .0048142 15.31 0.000 .0642708 .0831422
edad_ini_c~s | -.0243688 .0046722 -5.22 0.000 -.0335262 -.0152115
sex_enc | -.5041012 .0441559 -11.42 0.000 -.5906452 -.4175573
esc_rec | .2487896 .0265044 9.39 0.000 .196842 .3007373
sus_prin_mod | .2220002 .0182211 12.18 0.000 .1862874 .257713
fr_sus_prin | .0337664 .0166054 2.03 0.042 .0012204 .0663123
comp_biosoc | .2248986 .0301961 7.45 0.000 .1657152 .2840819
ten_viv | -.0250568 .0156652 -1.60 0.110 -.0557601 .0056465
dg_cie_10_~c | .0408707 .0186383 2.19 0.028 .0043403 .0774012
sud_sever~10 | -.1001643 .0422988 -2.37 0.018 -.1830685 -.0172602
macrozone | .2404059 .0243445 9.88 0.000 .1926916 .2881202
policonsumo | .0929098 .0486768 1.91 0.056 -.0024949 .1883146
n_off_vio | .3958711 .0364895 10.85 0.000 .324353 .4673891
n_off_acq | 1.090382 .0336267 32.43 0.000 1.024475 1.156289
n_off_sud | .3179083 .0358087 8.88 0.000 .2477245 .3880921
clas | .114424 .0286268 4.00 0.000 .0583166 .1705315
_cons | 5.828352 4.198461 1.39 0.165 -2.40048 14.05718
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP7
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_7
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18075.863 (not concave)
Iteration 2: log likelihood = -18074.131 (not concave)
Iteration 3: log likelihood = -18065.215 (not concave)
Iteration 4: log likelihood = -18061.798 (not concave)
Iteration 5: log likelihood = -18058.737 (not concave)
Iteration 6: log likelihood = -18055.853 (not concave)
Iteration 7: log likelihood = -18053.364 (not concave)
Iteration 8: log likelihood = -18051.265
Iteration 9: log likelihood = -18027.158 (backed up)
Iteration 10: log likelihood = -18005.933
Iteration 11: log likelihood = -17991.984
Iteration 12: log likelihood = -17974.566
Iteration 13: log likelihood = -17952.283
Iteration 14: log likelihood = -17951.276 (not concave)
Iteration 15: log likelihood = -17950.379 (not concave)
Iteration 16: log likelihood = -17950.37
Iteration 17: log likelihood = -17950.365
Iteration 18: log likelihood = -17950.365
Survival model Number of obs = 59,755
Log likelihood = -17950.365
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1302706 .0215693 6.04 0.000 .0879955 .1725457
edad_al_in~1 | .0790367 .0052145 15.16 0.000 .0688165 .0892569
edad_ini_c~s | -.0243046 .0046723 -5.20 0.000 -.0334622 -.015147
sex_enc | -.5038985 .0441556 -11.41 0.000 -.5904418 -.4173552
esc_rec | .2494012 .0265044 9.41 0.000 .1974534 .3013489
sus_prin_mod | .2220058 .0182286 12.18 0.000 .1862784 .2577333
fr_sus_prin | .0335933 .0166051 2.02 0.043 .0010479 .0661387
comp_biosoc | .2254081 .0302034 7.46 0.000 .1662106 .2846056
ten_viv | -.0254157 .015667 -1.62 0.105 -.0561225 .0052911
dg_cie_10_~c | .0408734 .0186421 2.19 0.028 .0043356 .0774112
sud_sever~10 | -.0997524 .0423 -2.36 0.018 -.1826588 -.0168459
macrozone | .2404287 .024342 9.88 0.000 .1927193 .2881382
policonsumo | .0951073 .0487023 1.95 0.051 -.0003476 .1905621
n_off_vio | .3945109 .0364912 10.81 0.000 .3229895 .4660324
n_off_acq | 1.088588 .0336344 32.37 0.000 1.022666 1.15451
n_off_sud | .3162357 .0358146 8.83 0.000 .2460404 .3864311
clas | .1148571 .0286259 4.01 0.000 .0587513 .1709628
_cons | 5.59626 4.229343 1.32 0.186 -2.693099 13.88562
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP8
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_8
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18075.903 (not concave)
Iteration 2: log likelihood = -18073.002 (not concave)
Iteration 3: log likelihood = -18064.75 (not concave)
Iteration 4: log likelihood = -18061.202 (not concave)
Iteration 5: log likelihood = -18058.174 (not concave)
Iteration 6: log likelihood = -18055.698 (not concave)
Iteration 7: log likelihood = -18053.594 (not concave)
Iteration 8: log likelihood = -18051.633 (not concave)
Iteration 9: log likelihood = -18049.82 (not concave)
Iteration 10: log likelihood = -18048.034
Iteration 11: log likelihood = -18029.74 (backed up)
Iteration 12: log likelihood = -18002.175
Iteration 13: log likelihood = -17980.049
Iteration 14: log likelihood = -17952.719
Iteration 15: log likelihood = -17948.754
Iteration 16: log likelihood = -17947.608 (not concave)
Iteration 17: log likelihood = -17947.608 (backed up)
Iteration 18: log likelihood = -17947.606
Iteration 19: log likelihood = -17947.606
Iteration 20: log likelihood = -17947.606 (backed up)
Survival model Number of obs = 59,755
Log likelihood = -17947.606
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1301788 .0215768 6.03 0.000 .087889 .1724686
edad_al_in~1 | .0830698 .0054628 15.21 0.000 .0723629 .0937768
edad_ini_c~s | -.0242941 .0046734 -5.20 0.000 -.0334538 -.0151344
sex_enc | -.5037363 .044155 -11.41 0.000 -.5902785 -.4171941
esc_rec | .2498123 .026502 9.43 0.000 .1978694 .3017553
sus_prin_mod | .2220603 .0182335 12.18 0.000 .1863232 .2577974
fr_sus_prin | .0334862 .0166045 2.02 0.044 .000942 .0660304
comp_biosoc | .2256518 .0302087 7.47 0.000 .1664438 .2848598
ten_viv | -.0256728 .0156682 -1.64 0.101 -.0563818 .0050363
dg_cie_10_~c | .0408292 .018645 2.19 0.029 .0042858 .0773727
sud_sever~10 | -.0995895 .0423001 -2.35 0.019 -.1824962 -.0166827
macrozone | .2404606 .02434 9.88 0.000 .192755 .2881662
policonsumo | .0967127 .0487168 1.99 0.047 .0012295 .1921959
n_off_vio | .3936155 .0364913 10.79 0.000 .3220938 .4651372
n_off_acq | 1.087179 .0336376 32.32 0.000 1.021251 1.153107
n_off_sud | .3151569 .0358168 8.80 0.000 .2449573 .3853565
clas | .1152424 .0286262 4.03 0.000 .0591361 .1713487
_cons | 5.519335 5.128446 1.08 0.282 -4.532234 15.5709
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP9
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_9
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18077.788 (not concave)
Iteration 2: log likelihood = -18072.064 (not concave)
Iteration 3: log likelihood = -18066.8 (not concave)
Iteration 4: log likelihood = -18054.402
Iteration 5: log likelihood = -18029.006 (backed up)
Iteration 6: log likelihood = -18003.959
Iteration 7: log likelihood = -17990.111
Iteration 8: log likelihood = -17971.554
Iteration 9: log likelihood = -17947.291
Iteration 10: log likelihood = -17944.351 (not concave)
Iteration 11: log likelihood = -17942.987
Iteration 12: log likelihood = -17942.948
Iteration 13: log likelihood = -17942.947
Survival model Number of obs = 59,755
Log likelihood = -17942.947
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1301544 .0215869 6.03 0.000 .0878449 .1724639
edad_al_in~1 | .0884168 .0057526 15.37 0.000 .0771419 .0996918
edad_ini_c~s | -.0242875 .0046749 -5.20 0.000 -.0334503 -.0151248
sex_enc | -.503899 .0441533 -11.41 0.000 -.5904379 -.4173601
esc_rec | .2504213 .0264998 9.45 0.000 .1984827 .3023599
sus_prin_mod | .2222191 .0182405 12.18 0.000 .1864683 .2579699
fr_sus_prin | .0332895 .0166039 2.00 0.045 .0007465 .0658325
comp_biosoc | .2258205 .0302111 7.47 0.000 .1666078 .2850332
ten_viv | -.0259655 .0156693 -1.66 0.098 -.0566768 .0047458
dg_cie_10_~c | .0409165 .0186501 2.19 0.028 .004363 .0774701
sud_sever~10 | -.0992754 .0423008 -2.35 0.019 -.1821834 -.0163675
macrozone | .2404345 .0243383 9.88 0.000 .1927323 .2881366
policonsumo | .098719 .0487332 2.03 0.043 .0032037 .1942342
n_off_vio | .3923337 .0364934 10.75 0.000 .320808 .4638593
n_off_acq | 1.085533 .0336423 32.27 0.000 1.019596 1.151471
n_off_sud | .3140495 .035819 8.77 0.000 .2438456 .3842535
clas | .1157444 .028627 4.04 0.000 .0596365 .1718523
_cons | 5.260914 2.724022 1.93 0.053 -.0780721 10.5999
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
***********
family RP10
***********
note; a delayed entry model is being fitted
Obtaining initial values
convergence not achieved
variables created: _rcs1_1 to _rcs1_10
Fitting full model:
Iteration 0: log likelihood = -18095.914 (not concave)
Iteration 1: log likelihood = -18083.341 (not concave)
Iteration 2: log likelihood = -18079.095 (not concave)
Iteration 3: log likelihood = -18067.784 (not concave)
Iteration 4: log likelihood = -18065.008 (not concave)
Iteration 5: log likelihood = -18061.084 (not concave)
Iteration 6: log likelihood = -18057.024 (not concave)
Iteration 7: log likelihood = -18050.671 (not concave)
Iteration 8: log likelihood = -18044.683 (not concave)
Iteration 9: log likelihood = -18039.196 (not concave)
Iteration 10: log likelihood = -18035.561
Iteration 11: log likelihood = -18011.856
Iteration 12: log likelihood = -17983.638
Iteration 13: log likelihood = -17962.308
Iteration 14: log likelihood = -17951.185
Iteration 15: log likelihood = -17939.864 (not concave)
Iteration 16: log likelihood = -17939.834
Iteration 17: log likelihood = -17939.74
Iteration 18: log likelihood = -17939.74
Survival model Number of obs = 59,755
Log likelihood = -17939.74
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_t: |
motivodeeg~3 | .1300112 .0215946 6.02 0.000 .0876865 .1723359
edad_al_in~1 | .0927599 .0060199 15.41 0.000 .0809612 .1045586
edad_ini_c~s | -.0242853 .0046773 -5.19 0.000 -.0334526 -.015118
sex_enc | -.5038711 .0441517 -11.41 0.000 -.5904069 -.4173353
esc_rec | .2512038 .026496 9.48 0.000 .1992725 .303135
sus_prin_mod | .2224494 .0182466 12.19 0.000 .1866867 .258212
fr_sus_prin | .0331595 .0166028 2.00 0.046 .0006187 .0657003
comp_biosoc | .2260753 .0302122 7.48 0.000 .1668605 .28529
ten_viv | -.0260899 .0156688 -1.67 0.096 -.0568002 .0046203
dg_cie_10_~c | .0409866 .0186535 2.20 0.028 .0044263 .0775469
sud_sever~10 | -.0991656 .0423005 -2.34 0.019 -.1820731 -.016258
macrozone | .2406702 .024337 9.89 0.000 .1929706 .2883698
policonsumo | .0999092 .0487411 2.05 0.040 .0043785 .1954399
n_off_vio | .3915739 .0364923 10.73 0.000 .3200503 .4630975
n_off_acq | 1.083966 .0336446 32.22 0.000 1.018023 1.149908
n_off_sud | .3127504 .0358205 8.73 0.000 .2425436 .3829572
clas | .1162043 .0286281 4.06 0.000 .0600942 .1723144
_cons | 5.345024 38.20365 0.14 0.889 -69.53275 80.2228
------------------------------------------------------------------------------
Warning: Baseline spline coefficients not shown - use ml display
.
. *rcs(time, df(3) orthog)
. estwrite _all using "${pathdata2}parmodels_m2_nov_22_2.sters", replace
(saving m2_1_cox1)
(saving m2_1_cox2)
(saving m2_1_cox3)
(saving m2_1_cox4)
(saving m2_1_cox5)
(saving m2_1_cox6)
(saving m2_1_cox7)
(saving m2_1_gom)
(saving m2_1_wei)
(saving m2_1_logl)
(saving m2_1_logn)
(saving m2_1_ggam)
(saving m2_1_rp1)
(saving m2_1_rp2)
(saving m2_1_rp3)
(saving m2_1_rp4)
(saving m2_1_rp5)
(saving m2_1_rp6)
(saving m2_1_rp7)
(saving m2_1_rp8)
(saving m2_1_rp9)
(saving m2_1_rp10)
(file parmodels_m2_nov_22_2.sters saved)
We obtained a summary of distributions by AICs and BICs.
. *estread "${pathdata2}parmodels_aic_bic_22_2.sters"
.
. *file:///G:/Mi%20unidad/Alvacast/SISTRAT%202019%20(github)/_supp_mstates/stata/1806.01615.pdf
. *rcs - restricted cubic splines on log hazard scale
. *rp - Royston-Parmar model (restricted cubic spline on log cumulative hazard scale)
. qui count if _d == 1
. // we count the amount of cases with the event in the strata
. //we call the estimates stored, and the results...
. estimates stat m2_1_*, n(`r(N)')
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | N ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
m2_1_cox1 | 5,144 . -18113.43 19 36264.86 36389.23
m2_1_cox2 | 5,144 . -18079.75 20 36199.49 36330.4
m2_1_cox3 | 5,144 . -18075.82 21 36193.65 36331.1
m2_1_cox4 | 5,144 . -18075.6 22 36195.2 36339.2
m2_1_cox5 | 5,144 . -18075.69 23 36197.39 36347.94
m2_1_cox6 | 5,144 . -18074.16 24 36196.32 36353.42
m2_1_cox7 | 5,144 . -18074.18 25 36198.36 36362
m2_1_gom | 5,144 . -17922.81 19 35883.63 36007.99
m2_1_wei | 5,144 . -18095.91 18 36227.83 36345.65
m2_1_logl | 5,144 . -18135.48 19 36308.96 36433.33
m2_1_logn | 5,144 . -18090.44 19 36218.89 36343.25
m2_1_ggam | 5,144 . -18070.36 14 36168.72 36260.35
m2_1_rp1 | 5,144 . -18095.91 18 36227.83 36345.65
m2_1_rp2 | 5,144 . -17968.8 20 35977.6 36108.51
m2_1_rp3 | 5,144 . -17969.1 21 35980.2 36117.66
m2_1_rp4 | 5,144 . -17967.83 22 35979.65 36123.66
m2_1_rp5 | 5,144 . -17958.54 23 35963.07 36113.62
m2_1_rp6 | 5,144 . -17954.12 24 35956.24 36113.33
m2_1_rp7 | 5,144 . -17950.36 25 35950.73 36114.37
m2_1_rp8 | 5,144 . -17947.61 26 35947.21 36117.4
m2_1_rp9 | 5,144 . -17942.95 27 35939.89 36116.62
m2_1_rp10 | 5,144 . -17939.74 28 35935.48 36118.76
-----------------------------------------------------------------------------
. //we store in a matrix de survival
. matrix stats_1=r(S)
.
.
. estimates clear
.
. ** to order AICs
. *https://www.statalist.org/forums/forum/general-stata-discussion/general/1665263-sorting-matrix-including-rownames
. mata :
------------------------------------------------- mata (type end to exit) -----------------------------------------------------
:
: void st_sort_matrix(
> //argumento de la matriz
> string scalar matname,
> //argumento de las columnas
> real rowvector columns
> )
> {
> string matrix rownames
> real colvector sort_order
> // defino una base
> //Y = st_matrix(matname)
> //[.,(1, 2, 3, 4, 6, 5)]
> //ordeno las columnas
> rownames = st_matrixrowstripe(matname) //[.,(1, 2, 3, 4, 6, 5)]
> sort_order = order(st_matrix(matname), (columns))
> st_replacematrix(matname, st_matrix(matname)[sort_order,.])
> st_matrixrowstripe(matname, rownames[sort_order,.])
> }
:
: end
-------------------------------------------------------------------------------------------------------------------------------
. //mata: mata drop st_sort_matrix()
.
. mata : st_sort_matrix("stats_1", 6)
. esttab matrix(stats_1) using "testreg_aic_bic_22_2.csv", replace
(output written to testreg_aic_bic_22_2.csv)
. esttab matrix(stats_1) using "testreg_aic_bic_22_2.html", replace
(output written to testreg_aic_bic_22_2.html)
.
| stats_1 | ||||||
| N | ll0 | ll | df | AIC | BIC | |
| m2_1_gom | 5144 | . | -17922.81 | 19 | 35883.63 | 36007.99 |
| m2_1_rp2 | 5144 | . | -17968.8 | 20 | 35977.6 | 36108.51 |
| m2_1_rp6 | 5144 | . | -17954.12 | 24 | 35956.24 | 36113.33 |
| m2_1_rp5 | 5144 | . | -17958.54 | 23 | 35963.07 | 36113.62 |
| m2_1_rp7 | 5144 | . | -17950.36 | 25 | 35950.73 | 36114.37 |
| m2_1_rp9 | 5144 | . | -17942.95 | 27 | 35939.89 | 36116.62 |
| m2_1_rp8 | 5144 | . | -17947.61 | 26 | 35947.21 | 36117.4 |
| m2_1_rp3 | 5144 | . | -17969.1 | 21 | 35980.2 | 36117.66 |
| m2_1_rp10 | 5144 | . | -17939.74 | 28 | 35935.48 | 36118.76 |
| m2_1_rp4 | 5144 | . | -17967.83 | 22 | 35979.65 | 36123.66 |
| m2_1_ggam | 5144 | . | -18070.36 | 14 | 36168.72 | 36260.35 |
| m2_1_cox2 | 5144 | . | -18079.75 | 20 | 36199.49 | 36330.4 |
| m2_1_cox3 | 5144 | . | -18075.82 | 21 | 36193.65 | 36331.1 |
| m2_1_cox4 | 5144 | . | -18075.6 | 22 | 36195.2 | 36339.2 |
| m2_1_logn | 5144 | . | -18090.44 | 19 | 36218.89 | 36343.25 |
| m2_1_wei | 5144 | . | -18095.91 | 18 | 36227.83 | 36345.65 |
| m2_1_rp1 | 5144 | . | -18095.91 | 18 | 36227.83 | 36345.65 |
| m2_1_cox5 | 5144 | . | -18075.69 | 23 | 36197.39 | 36347.94 |
| m2_1_cox6 | 5144 | . | -18074.16 | 24 | 36196.32 | 36353.42 |
| m2_1_cox7 | 5144 | . | -18074.18 | 25 | 36198.36 | 36362 |
| m2_1_cox1 | 5144 | . | -18113.43 | 19 | 36264.86 | 36389.23 |
| m2_1_logl | 5144 | . | -18135.48 | 19 | 36308.96 | 36433.33 |
. *reset time, only compatible with stteffects (same entry times)
. stset diff, failure(event ==1)
failure event: event == 1
obs. time interval: (0, diff]
exit on or before: failure
------------------------------------------------------------------------------
70,863 total observations
1 observation ends on or before enter()
------------------------------------------------------------------------------
70,862 observations remaining, representing
5,144 failures in single-record/single-failure data
302,812.78 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 0
last observed exit t = 10.75828
. *stset age_offending_imp, fail(event ==1) enter(edad_al_egres_imp)
.
. cap rm bsreg12.dta bsreg22.dta
.
. *count if missing(motivodeegreso_mod_imp_rec3, edad_al_ing_1, edad_ini_cons, dias_treat_imp_sin_na_1, esc_rec, sus_prin_mod,
> fr_sus_prin, comp_biosoc, ten_viv, dg_cie_10_rec, sud_severity_icd10, macrozone, policonsumo, n_off_vio, n_off_acq, n_off_sud
> , n_off_oth)
First we calculated the difference between those patients who did and did not complete baseline treatment, given that this analysis is restricted to 2 values.
. *Micki Hill & Paul C Lambert & Michael J Crowther, 2021. "Introducing stipw: inverse probability weighted parametric survival
> models," London Stata Conference 2021 15, Stata Users Group.
. *https://view.officeapps.live.com/op/view.aspx?src=http%3A%2F%2Ffmwww.bc.edu%2Frepec%2Fusug2021%2Fusug21_hill.pptx&wdOrigin=B
> ROWSELINK
.
. *Treatment variable should be a binary variable with values 0 and 1.
. gen motivodeegreso_mod_imp_rec2 = 0
. replace motivodeegreso_mod_imp_rec2 = 1 if strpos(motivodeegreso_mod_imp_rec,"Early")>0
(15,797 real changes made)
. replace motivodeegreso_mod_imp_rec2 = 1 if strpos(motivodeegreso_mod_imp_rec,"Late")>0
(35,781 real changes made)
.
. recode motivodeegreso_mod_imp_rec3 (1=0 "Tr Completion") (3=1 "Tr Non-completion (Late)") (2=2 "Tr Non-completion (Early)"),
> gen(caus_disch_mod_imp_rec)
(55066 differences between motivodeegreso_mod_imp_rec3 and caus_disch_mod_imp_rec)
. lab var caus_disch_mod_imp_rec "Baseline treatment outcome"
.
. global covs_3 "i.caus_disch_mod_imp_rec edad_al_ing_1 edad_ini_cons i.sex_enc i.esc_rec i.sus_prin_mod i.fr_sus_prin i.comp_b
> iosoc i.ten_viv i.dg_cie_10_rec i.sud_severity_icd10 i.macrozone i.policonsumo i.n_off_vio i.n_off_acq i.n_off_sud i.clas"
.
. global covs_3b "i.caus_disch_mod_imp_rec edad_al_ing_1 edad_ini_cons i.sex_enc i.esc_rec i.sus_prin_mod i.fr_sus_prin i.comp_
> biosoc i.origen_ingreso_mod numero_de_hijos_mod i.dg_cie_10_rec i.sud_severity_icd10 i.macrozone i.policonsumo i.n_off_vio i.
> n_off_acq i.n_off_sud i.clas"
.
. *______________________________________________
. *______________________________________________
. * ADJUSTED ROYSTON PARMAR - DF5, NO STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETION)
.
. stpm2 $covs_3 , scale(hazard) df(5) eform
Iteration 0: log likelihood = -17234.196
Iteration 1: log likelihood = -17231.695
Iteration 2: log likelihood = -17231.689
Iteration 3: log likelihood = -17231.689
Log likelihood = -17231.689 Number of obs = 59,755
-------------------------------------------------------------------------------------------------------------
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
--------------------------------------------+----------------------------------------------------------------
xb |
caus_disch_mod_imp_rec |
Tr Non-completion (Late) | 1.580846 .0801927 9.03 0.000 1.431232 1.7461
Tr Non-completion (Early) | 1.773658 .1068178 9.52 0.000 1.576183 1.995874
|
edad_al_ing_1 | .9656771 .0020826 -16.19 0.000 .9616039 .9697675
edad_ini_cons | .9753055 .0045823 -5.32 0.000 .9663655 .9843281
|
sex_enc |
Women | .5986703 .0265192 -11.58 0.000 .5488859 .6529701
|
esc_rec |
2-Completed high school or less | 1.347667 .0781147 5.15 0.000 1.202941 1.509804
3-Completed primary school or less | 1.580901 .0971921 7.45 0.000 1.401438 1.783346
|
sus_prin_mod |
Cocaine hydrochloride | 1.190785 .0789714 2.63 0.008 1.045641 1.356076
Cocaine paste | 2.013516 .105595 13.35 0.000 1.816834 2.231488
Marijuana | 1.335079 .1050888 3.67 0.000 1.144211 1.557787
Other | 1.827998 .2719367 4.05 0.000 1.365681 2.446821
|
fr_sus_prin |
2 to 3 days a week | 1.007899 .0813709 0.10 0.922 .8603925 1.180693
4 to 6 days a week | 1.04442 .0881621 0.51 0.607 .8851624 1.23233
Daily | 1.075197 .0856135 0.91 0.363 .9198359 1.256799
Less than 1 day a week | 1.039195 .1175102 0.34 0.734 .8326161 1.297028
|
comp_biosoc |
2-Moderate | 1.266761 .0998843 3.00 0.003 1.085369 1.478468
3-Severe | 1.496528 .124804 4.83 0.000 1.270862 1.762266
|
ten_viv |
Others | 1.05098 .1523573 0.34 0.732 .7910397 1.396338
Owner/Transferred dwellings/Pays Dividends | .8284898 .1030874 -1.51 0.131 .649193 1.057306
Renting | .8895818 .1129814 -0.92 0.357 .6935519 1.141019
Stays temporarily with a relative | .8337576 .1029156 -1.47 0.141 .6545925 1.061961
|
dg_cie_10_rec |
Diagnosis unknown (under study) | 1.104307 .0539795 2.03 0.042 1.003419 1.215338
With psychiatric comorbidity | 1.091941 .0419372 2.29 0.022 1.012763 1.177309
|
sud_severity_icd10 |
Hazardous consumption | .9542209 .041079 -1.09 0.276 .8770107 1.038229
|
macrozone |
North | 1.459505 .0587789 9.39 0.000 1.34873 1.579378
South | 1.523434 .0945671 6.78 0.000 1.348918 1.720529
|
1.policonsumo | 1.070096 .0523896 1.38 0.166 .9721866 1.177865
1.n_off_vio | 1.479257 .0541271 10.70 0.000 1.376885 1.589241
1.n_off_acq | 2.774692 .0941079 30.09 0.000 2.596241 2.965409
1.n_off_sud | 1.332665 .0478742 7.99 0.000 1.24206 1.429879
|
clas |
Rural | .968178 .0831156 -0.38 0.706 .8182419 1.145589
Urbana | 1.132403 .0686805 2.05 0.040 1.005485 1.275342
|
_rcs1 | 2.120386 .0233154 68.35 0.000 2.075178 2.166579
_rcs2 | 1.03444 .0083409 4.20 0.000 1.01822 1.050917
_rcs3 | 1.024283 .0067694 3.63 0.000 1.011101 1.037637
_rcs4 | 1.008095 .0048166 1.69 0.092 .9986986 1.01758
_rcs5 | 1.007675 .0035055 2.20 0.028 1.000828 1.014569
_cons | .0350692 .0075509 -15.56 0.000 .022996 .0534812
-------------------------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
.
. stpm2 $covs_3b , scale(hazard) df(5) eform
Iteration 0: log likelihood = -18606.277
Iteration 1: log likelihood = -18603.761
Iteration 2: log likelihood = -18603.756
Iteration 3: log likelihood = -18603.756
Log likelihood = -18603.756 Number of obs = 62,500
-----------------------------------------------------------------------------------------------------
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
------------------------------------+----------------------------------------------------------------
xb |
caus_disch_mod_imp_rec |
Tr Non-completion (Late) | 1.569941 .0766706 9.24 0.000 1.426637 1.72764
Tr Non-completion (Early) | 1.781134 .1025723 10.02 0.000 1.591026 1.993957
|
edad_al_ing_1 | .9625573 .0021803 -16.85 0.000 .9582935 .9668401
edad_ini_cons | .9757844 .0043571 -5.49 0.000 .967282 .9843616
|
sex_enc |
Women | .5545934 .0249293 -13.11 0.000 .5078235 .6056708
|
esc_rec |
2-Completed high school or less | 1.341292 .0752994 5.23 0.000 1.201537 1.497301
3-Completed primary school or less | 1.528273 .0912013 7.11 0.000 1.359579 1.717897
|
sus_prin_mod |
Cocaine hydrochloride | 1.160094 .0742901 2.32 0.020 1.023256 1.315232
Cocaine paste | 1.933162 .0973797 13.09 0.000 1.751421 2.133763
Marijuana | 1.288088 .0977557 3.34 0.001 1.110059 1.494669
Other | 1.752284 .2507087 3.92 0.000 1.323788 2.319479
|
fr_sus_prin |
2 to 3 days a week | .9988362 .0793161 -0.01 0.988 .8548731 1.167043
4 to 6 days a week | 1.044521 .0867684 0.52 0.600 .8875806 1.229211
Daily | 1.097092 .0858412 1.18 0.236 .9411123 1.278924
Less than 1 day a week | 1.076366 .1173255 0.68 0.500 .8693159 1.33273
|
comp_biosoc |
2-Moderate | 1.24577 .0959075 2.85 0.004 1.071289 1.448668
3-Severe | 1.478772 .1201811 4.81 0.000 1.261024 1.73412
|
origen_ingreso_mod |
Assisted Referral | 1.132939 .0562078 2.52 0.012 1.02796 1.248638
Other | 1.237856 .0751535 3.51 0.000 1.098984 1.394276
Justice Sector | 1.089302 .0616221 1.51 0.131 .9749796 1.217029
Health Sector | .9782757 .0384313 -0.56 0.576 .9057786 1.056575
|
numero_de_hijos_mod | 1.059249 .0139961 4.36 0.000 1.032169 1.087039
|
dg_cie_10_rec |
Diagnosis unknown (under study) | 1.088941 .050905 1.82 0.068 .993603 1.193426
With psychiatric comorbidity | 1.091049 .0402993 2.36 0.018 1.014855 1.172963
|
sud_severity_icd10 |
Hazardous consumption | .9597286 .0398832 -0.99 0.323 .8846577 1.04117
|
macrozone |
North | 1.475661 .0578781 9.92 0.000 1.366473 1.593574
South | 1.561551 .093145 7.47 0.000 1.389258 1.755212
|
1.policonsumo | 1.061669 .0490577 1.30 0.195 .9697431 1.162309
1.n_off_vio | 1.441362 .0509122 10.35 0.000 1.344952 1.544684
1.n_off_acq | 2.77654 .0901767 31.44 0.000 2.605305 2.959029
1.n_off_sud | 1.320371 .0454872 8.07 0.000 1.234161 1.412603
|
clas |
Rural | .9074739 .0761642 -1.16 0.247 .7698264 1.069733
Urbana | 1.103512 .0643768 1.69 0.091 .9842826 1.237185
|
_rcs1 | 2.118815 .0223299 71.25 0.000 2.075498 2.163036
_rcs2 | 1.036965 .008098 4.65 0.000 1.021214 1.052959
_rcs3 | 1.023772 .0064971 3.70 0.000 1.011117 1.036586
_rcs4 | 1.006088 .0045712 1.34 0.182 .9971687 1.015088
_rcs5 | 1.0073 .0033307 2.20 0.028 1.000794 1.01385
_cons | .0332125 .0056956 -19.85 0.000 .0237318 .0464807
-----------------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
.
. *______________________________________________
. *______________________________________________
. * INVERSE PROBABILITY WEIGHTED ADJUSTED ROYSTON PARMAR - DF5, NO STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETIO
> N)
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten
> _viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas), distribution(rp) df(5) genw(
> rpdf5_m_nostag_ten_viv) ipwtype(stabilised) vce(mestimation) eform
11108 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -35121.157
Iteration 1: log likelihood = -33474.525
Iteration 2: log likelihood = -33449.458
Iteration 3: log likelihood = -33449.44
Iteration 4: log likelihood = -33449.44
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -35121.157
Iteration 1: log likelihood = -35121.157
Fitting weighted survival model to obtain point estimates
Iteration 0: log pseudolikelihood = -18625.698
Iteration 1: log pseudolikelihood = -18622.913
Iteration 2: log pseudolikelihood = -18622.906
Iteration 3: log pseudolikelihood = -18622.906
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -18622.906 Number of obs = 59,755
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.665297 .0835877 10.16 0.000 1.509269 1.837455
_rcs1 | 2.073877 .0232869 64.96 0.000 2.028734 2.120024
_rcs2 | 1.040242 .009121 4.50 0.000 1.022518 1.058273
_rcs3 | 1.023716 .0072611 3.30 0.001 1.009583 1.038047
_rcs4 | 1.00922 .0048043 1.93 0.054 .9998477 1.01868
_rcs5 | 1.009227 .003495 2.65 0.008 1.0024 1.0161
_cons | .0354904 .0016903 -70.10 0.000 .0323273 .0389629
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ori
> gen_ingreso_mod numero_de_hijos_mod dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas
> ), distribution(rp) df(5) genw(rpdf5_m_nostag_or_ing_num_hij) ipwtype(stabilised) vce(mestimation) eform
8363 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -34725.723
Iteration 2: log likelihood = -34696.15
Iteration 3: log likelihood = -34696.127
Iteration 4: log likelihood = -34696.127
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -36548.788
Fitting weighted survival model to obtain point estimates
Iteration 0: log pseudolikelihood = -20117.963
Iteration 1: log pseudolikelihood = -20115.352
Iteration 2: log pseudolikelihood = -20115.347
Iteration 3: log pseudolikelihood = -20115.347
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -20115.347 Number of obs = 62,500
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.639443 .0789035 10.27 0.000 1.491864 1.801619
_rcs1 | 2.074982 .0222781 67.99 0.000 2.031774 2.119109
_rcs2 | 1.043215 .0086947 5.08 0.000 1.026312 1.060396
_rcs3 | 1.023483 .0068888 3.45 0.001 1.01007 1.037074
_rcs4 | 1.007252 .0045625 1.60 0.111 .9983494 1.016234
_rcs5 | 1.008696 .0033288 2.62 0.009 1.002192 1.015241
_cons | .037377 .001704 -72.09 0.000 .0341821 .0408706
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
.
. predict rmst03 in 1, at(motivodeegreso_mod_imp_rec2 0) rmst stdp tmax(3)
. predict rmst13 in 1, at(motivodeegreso_mod_imp_rec2 1) rmst stdp tmax(3)
. predictnl drmst= predict(rmst at(motivodeegreso_mod_imp_rec2 1) tmax(3))- predict(rmst at(motivodeegreso_mod_imp_rec2 1) tmax
> (3)) in 1, se(drmst_se)
Warning: prediction doesn't vary with respect to e(b).
(70,862 missing values generated)
Warning: prediction constant over observations; perhaps you meant to run nlcom.
.
. cap list rmst03 rmst13 drmst in 1
We used a gompertz distribution, assuming that baseline treatment outcome showed proportional hazards
. *______________________________________________
. *______________________________________________
. * INVERSE PROBABILITY WEIGHTED ADJUSTED GOMPERTZ - NO STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETION)
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten
> _viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas), distribution(gompertz) genw(
> gomp_m_nostag_ten_viv) ipwtype(stabilised) vce(mestimation)
11108 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -35121.157
Iteration 1: log likelihood = -33474.525
Iteration 2: log likelihood = -33449.458
Iteration 3: log likelihood = -33449.44
Iteration 4: log likelihood = -33449.44
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -35121.157
Iteration 1: log likelihood = -35121.157
Fitting weighted survival model to obtain point estimates
failure _d: event == 1
analysis time _t: diff
weight: [pweight=gomp_m_nostag_ten_viv]
Fitting constant-only model:
Iteration 0: log pseudolikelihood = -18959.069
Iteration 1: log pseudolikelihood = -18738.321
Iteration 2: log pseudolikelihood = -18729.984
Iteration 3: log pseudolikelihood = -18729.975
Iteration 4: log pseudolikelihood = -18729.975
Fitting full model:
Iteration 0: log pseudolikelihood = -18729.975
Iteration 1: log pseudolikelihood = -18652.414
Iteration 2: log pseudolikelihood = -18651.466
Iteration 3: log pseudolikelihood = -18651.466
Displaying weighted survival model with M-estimation standard errors
Gompertz PH regression Number of obs = 59,755
Wald chi2(1) = 102.94
Log pseudolikelihood = -18651.466 Prob > chi2 = 0.0000
---------------------------------------------------------------------------------------------
| M-estimation
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
motivodeegreso_mod_imp_rec2 | 1.66392 .0835046 10.15 0.000 1.508046 1.835906
_cons | .0172597 .0008842 -79.24 0.000 .0156108 .0190827
----------------------------+----------------------------------------------------------------
/gamma | -.1942119 .0100127 -19.40 0.000 -.2138364 -.1745873
---------------------------------------------------------------------------------------------
Note: _cons estimates baseline hazard.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ori
> gen_ingreso_mod numero_de_hijos_mod dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas
> ), distribution(gompertz) genw(gomp_m_nostag_or_ing_num_hij) ipwtype(stabilised) vce(mestimation)
8363 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -34725.723
Iteration 2: log likelihood = -34696.15
Iteration 3: log likelihood = -34696.127
Iteration 4: log likelihood = -34696.127
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -36548.788
Fitting weighted survival model to obtain point estimates
failure _d: event == 1
analysis time _t: diff
weight: [pweight=gomp_m_nostag_or_ing_num_hij]
Fitting constant-only model:
Iteration 0: log pseudolikelihood = -20478.705
Iteration 1: log pseudolikelihood = -20235.755
Iteration 2: log pseudolikelihood = -20226.682
Iteration 3: log pseudolikelihood = -20226.659
Iteration 4: log pseudolikelihood = -20226.659
Fitting full model:
Iteration 0: log pseudolikelihood = -20226.659
Iteration 1: log pseudolikelihood = -20147.658
Iteration 2: log pseudolikelihood = -20146.756
Iteration 3: log pseudolikelihood = -20146.756
Displaying weighted survival model with M-estimation standard errors
Gompertz PH regression Number of obs = 62,500
Wald chi2(1) = 105.12
Log pseudolikelihood = -20146.756 Prob > chi2 = 0.0000
---------------------------------------------------------------------------------------------
| M-estimation
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
motivodeegreso_mod_imp_rec2 | 1.637935 .0788273 10.25 0.000 1.490499 1.799954
_cons | .0180679 .0008824 -82.18 0.000 .0164187 .0198828
----------------------------+----------------------------------------------------------------
/gamma | -.1942989 .0096124 -20.21 0.000 -.2131389 -.1754589
---------------------------------------------------------------------------------------------
Note: _cons estimates baseline hazard.
.
. predict rmst03_c in 1, at(motivodeegreso_mod_imp_rec2 0) rmst stdp tmax(3)
option at() not allowed
r(198);
. predict rmst13_c in 1, at(motivodeegreso_mod_imp_rec2 1) rmst stdp tmax(3)
option at() not allowed
r(198);
. predictnl drmst_c= predict(rmst at(motivodeegreso_mod_imp_rec2 1) tmax(3))- predict(rmst at(motivodeegreso_mod_imp_rec2 1) tm
> ax(3)) in 1, se(drmst_c_se)
option rmst not allowed
predict(rmst at(motivodeegreso_mod_imp_rec2 1) tmax(3)) invalid
r(198);
.
. cap list rmst03_c rmst13_c drmst_c in 1
We used another model with only 2 degrees of freedom according to the lowest BIC
.
. *______________________________________________
. *______________________________________________
. * ADJUSTED ROYSTON PARMAR - DF2, NO STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETION)
.
. stpm2 $covs_3 , scale(hazard) df(2) eform
Iteration 0: log likelihood = -17255.265
Iteration 1: log likelihood = -17238.669
Iteration 2: log likelihood = -17238.625
Iteration 3: log likelihood = -17238.625
Log likelihood = -17238.625 Number of obs = 59,755
-------------------------------------------------------------------------------------------------------------
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
--------------------------------------------+----------------------------------------------------------------
xb |
caus_disch_mod_imp_rec |
Tr Non-completion (Late) | 1.580526 .0801744 9.02 0.000 1.430946 1.745741
Tr Non-completion (Early) | 1.77177 .106684 9.50 0.000 1.57454 1.993705
|
edad_al_ing_1 | .9657862 .0020824 -16.15 0.000 .9617133 .9698763
edad_ini_cons | .9751776 .0045818 -5.35 0.000 .9662386 .9841992
|
sex_enc |
Women | .5987102 .0265214 -11.58 0.000 .5489218 .6530145
|
esc_rec |
2-Completed high school or less | 1.346334 .0780354 5.13 0.000 1.201755 1.508306
3-Completed primary school or less | 1.579302 .0970888 7.43 0.000 1.400029 1.781532
|
sus_prin_mod |
Cocaine hydrochloride | 1.19043 .0789389 2.63 0.009 1.045345 1.355652
Cocaine paste | 2.008108 .1052966 13.30 0.000 1.811981 2.225463
Marijuana | 1.330003 .1046669 3.62 0.000 1.139897 1.551813
Other | 1.819524 .2706549 4.02 0.000 1.359381 2.435423
|
fr_sus_prin |
2 to 3 days a week | 1.006876 .0812875 0.08 0.932 .8595214 1.179494
4 to 6 days a week | 1.043037 .088046 0.50 0.618 .8839892 1.2307
Daily | 1.074868 .0855897 0.91 0.365 .9195503 1.256419
Less than 1 day a week | 1.038196 .1173982 0.33 0.740 .831814 1.295783
|
comp_biosoc |
2-Moderate | 1.266485 .0998575 3.00 0.003 1.08514 1.478135
3-Severe | 1.492968 .1244934 4.81 0.000 1.267861 1.758042
|
ten_viv |
Others | 1.048122 .1519439 0.32 0.746 .7888872 1.392543
Owner/Transferred dwellings/Pays Dividends | .8292786 .1031887 -1.50 0.132 .6498061 1.05832
Renting | .8899448 .1130312 -0.92 0.359 .6938293 1.141494
Stays temporarily with a relative | .8342702 .102983 -1.47 0.142 .6549886 1.062624
|
dg_cie_10_rec |
Diagnosis unknown (under study) | 1.105475 .0540238 2.05 0.040 1.004503 1.216596
With psychiatric comorbidity | 1.092734 .0419668 2.31 0.021 1.0135 1.178162
|
sud_severity_icd10 |
Hazardous consumption | .9548681 .0411061 -1.07 0.283 .877607 1.038931
|
macrozone |
North | 1.458862 .058752 9.38 0.000 1.348138 1.578681
South | 1.525214 .0947003 6.80 0.000 1.350454 1.722589
|
1.policonsumo | 1.065537 .0521407 1.30 0.195 .9680909 1.172792
1.n_off_vio | 1.484814 .0543324 10.80 0.000 1.382054 1.595215
1.n_off_acq | 2.789096 .094564 30.25 0.000 2.609778 2.980735
1.n_off_sud | 1.338274 .0480653 8.11 0.000 1.247307 1.435875
|
clas |
Rural | .9676282 .0830675 -0.38 0.701 .8177788 1.144936
Urbana | 1.130196 .0685454 2.02 0.044 1.003527 1.272854
|
_rcs1 | 2.114537 .0230575 68.67 0.000 2.069825 2.160216
_rcs2 | 1.039588 .0087519 4.61 0.000 1.022575 1.056884
_cons | .0351971 .0075782 -15.54 0.000 .02308 .0536755
-------------------------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
.
. stpm2 $covs_3b , scale(hazard) df(2) eform
Iteration 0: log likelihood = -18626.898
Iteration 1: log likelihood = -18610.3
Iteration 2: log likelihood = -18610.259
Iteration 3: log likelihood = -18610.259
Log likelihood = -18610.259 Number of obs = 62,500
-----------------------------------------------------------------------------------------------------
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
------------------------------------+----------------------------------------------------------------
xb |
caus_disch_mod_imp_rec |
Tr Non-completion (Late) | 1.569445 .0766446 9.23 0.000 1.42619 1.72709
Tr Non-completion (Early) | 1.778948 .1024245 10.00 0.000 1.589112 1.991461
|
edad_al_ing_1 | .9626527 .0021801 -16.81 0.000 .9583892 .9669352
edad_ini_cons | .9756694 .0043567 -5.52 0.000 .9671676 .9842459
|
sex_enc |
Women | .5547138 .024935 -13.11 0.000 .5079332 .605803
|
esc_rec |
2-Completed high school or less | 1.340196 .0752358 5.22 0.000 1.200559 1.496074
3-Completed primary school or less | 1.527171 .0911328 7.10 0.000 1.358604 1.716653
|
sus_prin_mod |
Cocaine hydrochloride | 1.159808 .0742634 2.32 0.021 1.023017 1.314889
Cocaine paste | 1.928451 .0971265 13.04 0.000 1.747181 2.128529
Marijuana | 1.283856 .0974146 3.29 0.001 1.106446 1.489713
Other | 1.745289 .2496887 3.89 0.000 1.318532 2.31017
|
fr_sus_prin |
2 to 3 days a week | .9979966 .0792489 -0.03 0.980 .8541553 1.166061
4 to 6 days a week | 1.043345 .0866718 0.51 0.610 .8865792 1.22783
Daily | 1.096827 .0858223 1.18 0.238 .9408815 1.278619
Less than 1 day a week | 1.075056 .1171828 0.66 0.507 .8682575 1.331108
|
comp_biosoc |
2-Moderate | 1.245305 .0958666 2.85 0.004 1.070898 1.448115
3-Severe | 1.475568 .1199068 4.79 0.000 1.258315 1.730331
|
origen_ingreso_mod |
Assisted Referral | 1.13159 .0561387 2.49 0.013 1.026741 1.247147
Other | 1.23398 .0749126 3.46 0.001 1.095553 1.389899
Justice Sector | 1.087033 .0614901 1.48 0.140 .9729557 1.214487
Health Sector | .9770014 .0383799 -0.59 0.554 .9046012 1.055196
|
numero_de_hijos_mod | 1.059284 .0139942 4.36 0.000 1.032208 1.08707
|
dg_cie_10_rec |
Diagnosis unknown (under study) | 1.090284 .0509523 1.85 0.064 .9948566 1.194865
With psychiatric comorbidity | 1.091903 .04033 2.38 0.017 1.015651 1.17388
|
sud_severity_icd10 |
Hazardous consumption | .9604382 .0399123 -0.97 0.331 .8853125 1.041939
|
macrozone |
North | 1.475441 .0578702 9.92 0.000 1.366268 1.593339
South | 1.563474 .0932775 7.49 0.000 1.390938 1.757412
|
1.policonsumo | 1.057931 .0488641 1.22 0.223 .9663662 1.158172
1.n_off_vio | 1.446422 .0510897 10.45 0.000 1.349675 1.550104
1.n_off_acq | 2.789555 .0905607 31.60 0.000 2.617588 2.972819
1.n_off_sud | 1.325195 .0456445 8.17 0.000 1.238686 1.417745
|
clas |
Rural | .906769 .0761044 -1.17 0.244 .7692296 1.068901
Urbana | 1.101454 .0642563 1.66 0.098 .9824473 1.234876
|
_rcs1 | 2.113744 .0221173 71.53 0.000 2.070836 2.15754
_rcs2 | 1.042234 .0084581 5.10 0.000 1.025788 1.058945
_cons | .0333609 .0057207 -19.83 0.000 .0238383 .0466876
-----------------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
.
. *______________________________________________
. *______________________________________________
. * INVERSE PROBABILITY WEIGHTED ADJUSTED ROYSTON PARMAR - DF2, NO STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETIO
> N)
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten
> _viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas), distribution(rp) df(2) genw(
> rpdf2_m_nostag_ten_viv) ipwtype(stabilised) vce(mestimation) eform
11108 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -35121.157
Iteration 1: log likelihood = -33474.525
Iteration 2: log likelihood = -33449.458
Iteration 3: log likelihood = -33449.44
Iteration 4: log likelihood = -33449.44
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -35121.157
Iteration 1: log likelihood = -35121.157
Fitting weighted survival model to obtain point estimates
Iteration 0: log pseudolikelihood = -18647.26
Iteration 1: log pseudolikelihood = -18630.822
Iteration 2: log pseudolikelihood = -18630.775
Iteration 3: log pseudolikelihood = -18630.775
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -18630.775 Number of obs = 59,755
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.664996 .0835719 10.16 0.000 1.508997 1.837121
_rcs1 | 2.066911 .023429 64.05 0.000 2.021498 2.113345
_rcs2 | 1.046111 .0100606 4.69 0.000 1.026577 1.066016
_cons | .0355151 .001691 -70.10 0.000 .0323507 .038989
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ori
> gen_ingreso_mod numero_de_hijos_mod dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas
> ), distribution(rp) df(2) genw(rpdf2_m_nostag_or_ing_num_hij) ipwtype(stabilised) vce(mestimation) eform
8363 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -34725.723
Iteration 2: log likelihood = -34696.15
Iteration 3: log likelihood = -34696.127
Iteration 4: log likelihood = -34696.127
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -36548.788
Fitting weighted survival model to obtain point estimates
Iteration 0: log pseudolikelihood = -20139.24
Iteration 1: log pseudolikelihood = -20122.77
Iteration 2: log pseudolikelihood = -20122.725
Iteration 3: log pseudolikelihood = -20122.725
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -20122.725 Number of obs = 62,500
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.639117 .0788877 10.27 0.000 1.491569 1.801262
_rcs1 | 2.068827 .022426 67.07 0.000 2.025336 2.113251
_rcs2 | 1.049317 .0095814 5.27 0.000 1.030705 1.068265
_cons | .0374018 .0017046 -72.10 0.000 .0342057 .0408967
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
.
. predict rmst03_b in 1, at(motivodeegreso_mod_imp_rec2 0) rmst stdp tmax(3)
. predict rmst13_b in 1, at(motivodeegreso_mod_imp_rec2 1) rmst stdp tmax(3)
. predictnl drmst_b= predict(rmst at(motivodeegreso_mod_imp_rec2 1) tmax(3))- predict(rmst at(motivodeegreso_mod_imp_rec2 1) tm
> ax(3)) in 1, se(drmst_b_se)
Warning: prediction doesn't vary with respect to e(b).
(70,862 missing values generated)
Warning: prediction constant over observations; perhaps you meant to run nlcom.
.
. cap list rmst03_b rmst13_b drmst_b in 1
Staggered entry
. *______________________________________________
. *______________________________________________
. * ADJUSTED ROYSTON PARMAR - DF5, STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETION)
.
. stset age_offending_imp, fail(event ==1) enter(edad_al_egres_imp)
failure event: event == 1
obs. time interval: (0, age_offending_imp]
enter on or after: time edad_al_egres_imp
exit on or before: failure
------------------------------------------------------------------------------
70,863 total observations
1 observation ends on or before enter()
------------------------------------------------------------------------------
70,862 observations remaining, representing
5,144 failures in single-record/single-failure data
302,812.78 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 10.95068
last observed exit t = 90.65027
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten
> _viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud), distribution(rp) df(5) genw(rpdf5
> _m_stag_ten_viv) ipwtype(stabilised) vce(mestimation) eform
11106 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -35121.799
Iteration 1: log likelihood = -33484.63
Iteration 2: log likelihood = -33459.824
Iteration 3: log likelihood = -33459.806
Iteration 4: log likelihood = -33459.806
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -35121.799
Iteration 1: log likelihood = -35121.799
Fitting weighted survival model to obtain point estimates
note: delayed entry models are being fitted
Iteration 0: log pseudolikelihood = -5542.8717
Iteration 1: log pseudolikelihood = -5523.0629
Iteration 2: log pseudolikelihood = -5514.988
Iteration 3: log pseudolikelihood = -5511.8346
Iteration 4: log pseudolikelihood = -5511.307
Iteration 5: log pseudolikelihood = -5511.1369
Iteration 6: log pseudolikelihood = -5511.1269
Iteration 7: log pseudolikelihood = -5511.1216
Iteration 8: log pseudolikelihood = -5511.1214
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -5511.1214 Number of obs = 59,757
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.610324 .0817913 9.38 0.000 1.457737 1.778883
_rcs1 | 1.158443 .0647043 2.63 0.008 1.03832 1.292463
_rcs2 | 1.051606 .0261534 2.02 0.043 1.001575 1.104136
_rcs3 | 1.000906 .0035305 0.26 0.797 .99401 1.007849
_rcs4 | 1.002826 .0009749 2.90 0.004 1.000917 1.004739
_rcs5 | .9997684 .0009992 -0.23 0.817 .9978119 1.001729
_cons | .6834873 .2467749 -1.05 0.292 .3368231 1.386944
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
. estimates store df5_stipw
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ori
> gen_ingreso_mod numero_de_hijos_mod dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud), di
> stribution(rp) df(5) genw(rpdf5_m_stag_or_ing_num_hij) ipwtype(stabilised) vce(mestimation) eform
8361 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -36549.421
Iteration 1: log likelihood = -34737.642
Iteration 2: log likelihood = -34708.396
Iteration 3: log likelihood = -34708.373
Iteration 4: log likelihood = -34708.373
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -36549.421
Iteration 1: log likelihood = -36549.421
Fitting weighted survival model to obtain point estimates
note: delayed entry models are being fitted
Iteration 0: log pseudolikelihood = -5918.6569
Iteration 1: log pseudolikelihood = -5903.8654
Iteration 2: log pseudolikelihood = -5886.952
Iteration 3: log pseudolikelihood = -5886.586
Iteration 4: log pseudolikelihood = -5886.3212
Iteration 5: log pseudolikelihood = -5886.3155
Iteration 6: log pseudolikelihood = -5886.3151
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -5886.3151 Number of obs = 62,502
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.578604 .0768661 9.38 0.000 1.434915 1.736682
_rcs1 | 1.187069 .0588327 3.46 0.001 1.077183 1.308165
_rcs2 | 1.062082 .0250486 2.55 0.011 1.014105 1.112329
_rcs3 | .9991374 .0046417 -0.19 0.853 .9900811 1.008276
_rcs4 | 1.002728 .0010768 2.54 0.011 1.000619 1.00484
_rcs5 | 1.000176 .0009839 0.18 0.858 .9982496 1.002106
_cons | .6226638 .1695861 -1.74 0.082 .3651102 1.061899
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
. estimates store df5_stipw2
. *______________________________________________
. *______________________________________________
. * INVERSE PROBABILITY WEIGHTED ADJUSTED GOMPERTZ - STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETION)
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten
> _viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud), distribution(gompertz) genw(gomp_
> m_stag_ten_viv) ipwtype(stabilised) vce(mestimation)
11106 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -35121.799
Iteration 1: log likelihood = -33484.63
Iteration 2: log likelihood = -33459.824
Iteration 3: log likelihood = -33459.806
Iteration 4: log likelihood = -33459.806
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -35121.799
Iteration 1: log likelihood = -35121.799
Fitting weighted survival model to obtain point estimates
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
weight: [pweight=gomp_m_stag_ten_viv]
Fitting constant-only model:
Iteration 0: log pseudolikelihood = -6286.1603
Iteration 1: log pseudolikelihood = -5654.2431
Iteration 2: log pseudolikelihood = -5587.4209
Iteration 3: log pseudolikelihood = -5587.2593
Iteration 4: log pseudolikelihood = -5587.2593
Fitting full model:
Iteration 0: log pseudolikelihood = -5587.2593
Iteration 1: log pseudolikelihood = -5519.7285
Iteration 2: log pseudolikelihood = -5519.0123
Iteration 3: log pseudolikelihood = -5519.0122
Displaying weighted survival model with M-estimation standard errors
Gompertz PH regression Number of obs = 59,757
Wald chi2(1) = 88.15
Log pseudolikelihood = -5519.0122 Prob > chi2 = 0.0000
---------------------------------------------------------------------------------------------
| M-estimation
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
motivodeegreso_mod_imp_rec2 | 1.610994 .0818218 9.39 0.000 1.45835 1.779616
_cons | .1194776 .0100089 -25.36 0.000 .1013864 .1407971
----------------------------+----------------------------------------------------------------
/gamma | -.0663733 .0019544 -33.96 0.000 -.0702039 -.0625428
---------------------------------------------------------------------------------------------
Note: _cons estimates baseline hazard.
. estimates store gomp_stipw
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ori
> gen_ingreso_mod numero_de_hijos_mod dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud clas
> ), distribution(gompertz) genw(gomp_m_stag_or_ing_num_hij) ipwtype(stabilised) vce(mestimation)
8363 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -34725.723
Iteration 2: log likelihood = -34696.15
Iteration 3: log likelihood = -34696.127
Iteration 4: log likelihood = -34696.127
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -36548.788
Iteration 1: log likelihood = -36548.788
Fitting weighted survival model to obtain point estimates
failure _d: event == 1
analysis time _t: age_offending_imp
enter on or after: time edad_al_egres_imp
weight: [pweight=gomp_m_stag_or_ing_num_hij]
Fitting constant-only model:
Iteration 0: log pseudolikelihood = -6696.1223
Iteration 1: log pseudolikelihood = -6025.7735
Iteration 2: log pseudolikelihood = -5957.0954
Iteration 3: log pseudolikelihood = -5956.953
Iteration 4: log pseudolikelihood = -5956.953
Fitting full model:
Iteration 0: log pseudolikelihood = -5956.953
Iteration 1: log pseudolikelihood = -5890.7529
Iteration 2: log pseudolikelihood = -5890.1189
Iteration 3: log pseudolikelihood = -5890.1188
Displaying weighted survival model with M-estimation standard errors
Gompertz PH regression Number of obs = 62,500
Wald chi2(1) = 86.46
Log pseudolikelihood = -5890.1188 Prob > chi2 = 0.0000
---------------------------------------------------------------------------------------------
| M-estimation
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
motivodeegreso_mod_imp_rec2 | 1.573035 .0766373 9.30 0.000 1.429777 1.730647
_cons | .1204473 .0096851 -26.32 0.000 .1028851 .1410075
----------------------------+----------------------------------------------------------------
/gamma | -.0650709 .001865 -34.89 0.000 -.0687263 -.0614156
---------------------------------------------------------------------------------------------
Note: _cons estimates baseline hazard.
. estimates store gomp_stipw2
Given that the model with 2 degrees of freedom did not converge, we calculated the estimates with the second model with best AIC (3 degrees of freedom).
.
. *______________________________________________
. *______________________________________________
. * ADJUSTED ROYSTON PARMAR - DF3, STAGGERED ENTRY, BINARY TREATMENT (1-DROPOUT VS. 0-COMPLETION)
.
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ten
> _viv dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud), distribution(rp) df(3) genw(rpdf3
> _m_stag_ten_viv) ipwtype(stabilised) vce(mestimation) eform
11106 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -35121.799
Iteration 1: log likelihood = -33484.63
Iteration 2: log likelihood = -33459.824
Iteration 3: log likelihood = -33459.806
Iteration 4: log likelihood = -33459.806
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -35121.799
Iteration 1: log likelihood = -35121.799
Fitting weighted survival model to obtain point estimates
note: delayed entry models are being fitted
Iteration 0: log pseudolikelihood = -5610.0407 (not concave)
Iteration 1: log pseudolikelihood = -5539.2718
Iteration 2: log pseudolikelihood = -5529.4155
Iteration 3: log pseudolikelihood = -5519.5292
Iteration 4: log pseudolikelihood = -5518.0052
Iteration 5: log pseudolikelihood = -5516.3065
Iteration 6: log pseudolikelihood = -5515.8985
Iteration 7: log pseudolikelihood = -5515.7518
Iteration 8: log pseudolikelihood = -5515.423
Iteration 9: log pseudolikelihood = -5515.3463
Iteration 10: log pseudolikelihood = -5515.3098
Iteration 11: log pseudolikelihood = -5515.2666 (not concave)
Iteration 12: log pseudolikelihood = -5515.2646
Iteration 13: log pseudolikelihood = -5515.2559
Iteration 14: log pseudolikelihood = -5515.2505
Iteration 15: log pseudolikelihood = -5515.2501
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -5515.2501 Number of obs = 59,757
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.611422 .0818826 9.39 0.000 1.458668 1.780173
_rcs1 | 1.086826 .036706 2.47 0.014 1.017213 1.161203
_rcs2 | 1.025681 .012305 2.11 0.035 1.001845 1.050084
_rcs3 | 1.002548 .0007403 3.45 0.001 1.001098 1.004
_cons | 1.176211 .4660035 0.41 0.682 .5410628 2.556955
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
. estimates store df3_stipw
. stipw (logit motivodeegreso_mod_imp_rec2 edad_al_ing_1 edad_ini_cons sex_enc esc_rec sus_prin_mod fr_sus_prin comp_biosoc ori
> gen_ingreso_mod numero_de_hijos_mod dg_cie_10_rec sud_severity_icd10 macrozone policonsumo n_off_vio n_off_acq n_off_sud), di
> stribution(rp) df(3) genw(rpdf3_m_stag_or_ing_num_hij) ipwtype(stabilised) vce(mestimation) eform
8361 observations have missing treatment and/or missing confounder values and/or _st = 0.
These observations are excluded from the analysis, see variable _stipw_flag
Fitting logistic regression to obtain denominator for weights
Iteration 0: log likelihood = -36549.421
Iteration 1: log likelihood = -34737.642
Iteration 2: log likelihood = -34708.396
Iteration 3: log likelihood = -34708.373
Iteration 4: log likelihood = -34708.373
Fitting second logistic regression with no confounders to obtain numerator for stabilised weights
Iteration 0: log likelihood = -36549.421
Iteration 1: log likelihood = -36549.421
Fitting weighted survival model to obtain point estimates
note: delayed entry models are being fitted
Iteration 0: log pseudolikelihood = -5998.0505
Iteration 1: log pseudolikelihood = -5936.2125
Iteration 2: log pseudolikelihood = -5894.7933
Iteration 3: log pseudolikelihood = -5891.7206
Iteration 4: log pseudolikelihood = -5891.1581
Iteration 5: log pseudolikelihood = -5890.7076
Iteration 6: log pseudolikelihood = -5890.5856
Iteration 7: log pseudolikelihood = -5890.5416
Iteration 8: log pseudolikelihood = -5890.5181
Iteration 9: log pseudolikelihood = -5890.5176
Iteration 10: log pseudolikelihood = -5890.5149
Iteration 11: log pseudolikelihood = -5890.5148
Displaying weighted survival model with M-estimation standard errors
Log pseudolikelihood = -5890.5148 Number of obs = 62,502
---------------------------------------------------------------------------------------------
| M-estimation
| exp(b) Std. Err. z P>|z| [95% Conf. Interval]
----------------------------+----------------------------------------------------------------
xb |
motivodeegreso_mod_imp_rec2 | 1.579325 .0769329 9.38 0.000 1.435514 1.737543
_rcs1 | 1.118239 .0351403 3.56 0.000 1.051444 1.189278
_rcs2 | 1.035826 .0125568 2.90 0.004 1.011505 1.060732
_rcs3 | 1.002315 .0009006 2.57 0.010 1.000552 1.004082
_cons | .9316145 .2545062 -0.26 0.795 .5453784 1.591383
---------------------------------------------------------------------------------------------
Note: Estimates are transformed only in the first equation.
. estimates store df3_stipw2
. qui count if _d == 1
. // we count the amount of cases with the event in the strata
. //we call the estimates stored, and the results...
. estimates stat df5_stipw gomp_stipw df3_stipw, n(`r(N)')
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | N ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
df5_stipw | 5,144 . -5511.121 7 11036.24 11082.06
gomp_stipw | 5,144 -5587.259 -5519.012 3 11044.02 11063.66
df3_stipw | 5,144 . -5515.25 5 11040.5 11073.23
-----------------------------------------------------------------------------
. //we store in a matrix de survival
. matrix stats_stipw=r(S)
.
. estwrite df5_stipw gomp_stipw df3_stipw df5_stipw2 gomp_stipw2 df3_stipw2 using "${pathdata2}parmodels_m2_stipw_22.sters", re
> place
(saving df5_stipw)
(saving gomp_stipw)
(saving df3_stipw)
(saving df5_stipw2)
(saving gomp_stipw2)
(saving df3_stipw2)
(file parmodels_m2_stipw_22.sters saved)
Saved at= 17:03:52 16 Feb 2023