stata - Logit model (backed up) Warning -
i trying run logit identifying probability of being tribal versus non-tribal on subsample of adult women only. have 5 rounds of data , independent variable includes years of education, age, marital status, rural residence, children under 5, household size, age of household head , monthly per capita income. have included set of zone dummies zone1-zone6 divides country 6 unique zone-with each zone consisting of group of states.
this command use
foreach r in 1 2 3 4 5 { logit scst age yrs_ed rural marital n_mpce ch_under5 fhead headage heademp hhsize i.zone [aw=hhwt] /* */ if (keep_agerc==1) & (sex==2) & (round==`r'), iterate (100) predict phatscst`r' if e(sample), pr sum scst [aw=hhwt] if (keep_agerc==1) & (sex==2) & (round==`r') gen pbarscst`r'=r(mean) gen scstwt`r'= (phatscst`r'/(1-phatscst`r')) * ((1-pbarscst`r')/ pbarscst`r') * (hhwt) if scst==0 & round==`r' }
the issue run rounds 1,2 , 3 logit converges perfectly, rounds 4 , 5, after 6 iterations, model starts saying (backed up) this
iteration 0: log pseudolikelihood = -1.683e+08 iteration 1: log pseudolikelihood = -1.557e+08 iteration 2: log pseudolikelihood = -1.545e+08 iteration 3: log pseudolikelihood = -1.545e+08 iteration 4: log pseudolikelihood = -1.545e+08 (backed up) iteration 5: log pseudolikelihood = -1.545e+08 (backed up) iteration 6: log pseudolikelihood = -1.545e+08 (backed up) iteration 7: log pseudolikelihood = -1.545e+08 (backed up) iteration 8: log pseudolikelihood = -1.545e+08 (backed up) iteration 9: log pseudolikelihood = -1.545e+08 (backed up) iteration 10: log pseudolikelihood = -1.545e+08 (backed up) . . .<output omitted> warning: convergence not achieved
i tried introducing variables sequentially , seems monthly per capita income (n_mpce) problem variable. cannot remove model since it's measure of socioeconomic status have. tried converting daily per capita income reduce scale- doesn't seem work either.
my question is
- what stata mean (backed up)- how different (not concave) warning.
- if restrict iterations 100 , broadly results want (for reweighting)- how big of issue non-convergence of model.
- why non-convergence happening in round 4 , 5 , not in earlier rounds- unit of measurement of mpce has been same in rounds.
also posted at: http://www.statalist.org/forums/forum/general-stata-discussion/general/1109966-logit-model-backed-up-warning
Comments
Post a Comment