Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: LPM probit logit for Panel data


From   Chirantan Chatterjee <chirantan@cmu.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: LPM probit logit for Panel data
Date   Sat, 5 Apr 2008 03:55:55 -0400 (EDT)

Hi, 

I am trying to run a model of the exporting decision of some firms on some RHS variables. The LHS variable is a dummy variable taking values of 0 or 1 depending on whether a firm is in an export market in a certain year or not. The data is a 16 year unbalanced panel of 320 firms. 

My question relates to running this on either of the above binary choice models:

*  For LPM, i want to include firm and year fixed effects, in that case what is the right way:

xtreg Y X1 X2 X3 year dummies, fe
or
regress Y X1 X2 X3 year dummies firmdummy1-firmdummy2

I find that standard errors are different if i do the above two, though the t-stat and the point estimates are much the same.

*  For Probit, can anyone suggest how to include firm fixed effects if at all without including all the firm dummies in the above regressions. Is there any short cut?

*  For logit, i am using the clogit command thus:

clogit Y X1 X2 X3 time dummies, group(firmid)

- Is this correct to incorporate firm and year fixed effects for a panel logit. 


Thanks, Chirantan.
Doctoral Student, Heinz School
Carnegie Mellon, PA. 




-- 


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index