[kusum shekhawat]

Hi,
I have been going through various research papers and articles
which suggests that mixed model is a better option when we are dealing with healthcare cost, specially if we are dealing with zeroes.
i used the following command for my model

"twopm total_cost age i.age_grp i.sex i.comorb_cat ib2.health_insurance i.wealth_tertile i.facility1 i.level1 i.treatment1 i.flu1 ib2.sample_type_final ib2.Site, ///
firstpart(logit, nolog) secondpart(glm, family(gamma) link(log) nolog)"

but getting different number of observations in the first part, i dont know why

i have attached the logfile of the output and reference articles for choosing two part model(Hoping i chose right)


https://clas.ucdenver.edu/marcelo-pe...perraillon.pdf
https://academic.oup.com/intqhc/arti.../3/331/1792904
https://journals.sagepub.com/doi/pdf...867X1501500102

[Nick Cox]

Two-part models can be a great idea, but I think you need to start a new thread on that. In particular, a title mentioning two-part models will attract experts not so interested in discussing transformations.

Links can be helpful, but please try to make questions self-contained. .

SMCL attachments should be readable with a little effort by members here but it's much, much better to post output as CODE, exactly as explained in the FAQ Advice.

[kusum shekhawat]

sure Nick Cox
i have started a new thread named two part models and included the code snippet in that, thanks for the help