Dear Stata users,
Apologies if you have already received this email - I was not sure
whether it got sent as I didn't receive a copy:
I have a few questions regarding the heckman estimator:
1) Is there a way to accomodate heteroskedastic errors into the heckman
ml estimator?
2) My second problem is similar to others I have seen in the archives
but I haven't seen a solution - I am getting very different results
when I use the ml over the two-step estimator. I thought the 2-step is
consistent but that the ml is more efficient. However, I am getting
completely different coefficients and s.e. from both methods. Unlike
previous queries in the archive my log likelihood seems to be maximising
ok (no concave functions at the last iteration - although there are some
at the beginning which I thought was fine).
Any suggestions on how to proceed on either query is much appreciated.
Thank you,
Lupin
Lupin Rahman
EOPP, STICERD
London School of Economics
Houghton Street, London
WC2A 2AE
Tel: +44 207 955 6677
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
