I now have an econometrics model: y1=x1b1+e1 (y1 is a binary variable), and y2=x2b2+e2 (y2 is observed while y1=1). And I have two problems in this model, one is x2 is correlated with e2 which I already used IV method to eliminate, and the other is the selection problem which I wish to use heckman 2-step estimation. However, I don't know how to use heckman and IV together with the command in Stata.
"heckman y2 x2, col select(y1=x1) first" works, yet I couldn't use IV here: "ivreg y2 (x2=Z2+e) if y1==1, first" also works, yet I couldn't add the heckman estimation in.
Is there anyway to work this problem out in Stata? Thanks a lot!
Sailu (Lulu) LI
Economics Dept., Boston College
140 Commonwealth Ave.
Chestnut Hill, MA 02467
Cell: (617)990-7033
*
* 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/
