Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 72771

Using CMP to fit Craggit

$
0
0
Hi everyone,

I am learning about cmp command, and trying to use it instead of craggit command. However, when I compare the output with craggit (no matter how I do it), I get different results. It looks like the output of cmp for probit is fine, but the truncated regression part is just OLS.

Given is the setup I use for cmp and craggit. For cmp, It does not work, so I add nonrtolerance option and results are somewhat similar. But still, what am I doing wrong?

webuse laborsup, clear
replace fem_inc = fem_inc - 10
gen selection=(fem_inc>0)
craggit selection kids male_educ, second(fem_inc kids male_educ)
cmp (selection=kids male_educ)(fem_inc=kids male_educ, trunc(0 .)), indicators($cmp_probit $cmp_trunc)

Viewing all articles
Browse latest Browse all 72771

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>