Dear Sirs,
I am a ph d student at the University of Cambridge (Serial Number:
81990517613) and I am writing my ph d dissertation on the political
economics of U.S. regulated markets. In one of my recent papers I have used
a random effects with one interaction term, i. e.
y(i,t) = x1(i,t)b1 + x2(i,t)b1 + x1(i,t)*x2(i,t)b3 + bZ + e.
After a referee's remark, I discovered that stata report uncorrect
estimates of both the interaction terms (b1, b2 and b3 above) and of the
relative standard errors when nonlinear index models are estimated (see Ai
and Norton (2003)).
Norton, Wand and Ai (2004) - Stata Journal - have written a program in
Stata able to fix the problem for logit and probit model (Inteff command in
Stata), however, as far as I know, there isn't a command doing the same for
random or fixed effect logit.
May you help me in understanding how to fix this problem. The issue is
absolutely relevant for me.
I thank you for your time and consideration and I look forward to hearing
further from you.
Bests,
Carmine Guerriero.
*
* 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/
