Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Gustavo Sanchez" <gsanchez@stata.com> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | RE: st: RE: Error? xtdpdsys assigns explanatory power to fixed effects |
Date | Thu, 10 Nov 2011 10:14:03 -0600 |
Lara <larasusannakrugman@gmail.com> has a follow-up question regarding her estimation results using -xtabond- and -xtdpdsys-: > I estimate the effects of a certain variable W on the dependent variable Y. > I know that the effect is different depending on the type of country. > Therefore I created dummies for mutually exclusive groups of countries (a & b). > Now I interact the variable W with these dummies (W*a, W*b) to estimate the > effect of W with regard to the type of country. By estimating subsamples > (a==1 or b==1) I know that W has a positive effect for a-countries but a negative > effect for b-countries. If I estimate the whole sample using XTDPDSYS and > including the interaction terms W*a and W*b I falsely obtain positive coefficients > for both. (If I use XTABOND I find a positive and a negative coefficient as expected.) > Is this a regression that cannot be done using XTDPDSYS as it estimates explanatory > power for the 'zeros' in (W*a) for the b-countries?? As I explained in my previous response, the two estimators (Arellano/Bond and Blundell-Bond/Arellano-Bover) handle the time invariant dummies in a different way. This is a theoretical issue that is reflected in the results she is obtaining with the Stata commands she is using for her dynamic panel data estimation. However, without looking at Lara's log containing the actual command lines and the regression results, it is not straightforward to determine whether the differences she is reporting are associated to the way the two estimators handle the dummy variables included in her model, or whether there is any other particular problem in her code. Lara can send her log and her dataset to my private email, so that I can provide more specific comments and hints on the results she is getting. --Gustavo gsanchez@stata.com * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/