Re: st: panel data: xtrege, re vs. xtreg, fe vs. regress...cluster

[nicola.baldini2@unibo.it]

Looking at   
http://www.kellogg.northwestern.edu/faculty/petersen/htm/papers/se/se_programming.htm
can provide you some suggestions about how to correct the standard errors in panel data analysis.

P.S. As reported, -xtreg- (GLS) can improve efficiency over -regress- not only for the standard errors, but also for the estimation of the coefficients
P.P.S. are you sure that a linear regression command is useful for a y that is bounded between -1 and 1? what about interval/censored regression? this solution would "solve" the problems, since only random effects are available

Nicola

At 02.33 12/03/2007 -0400, Michael Pfarrer wrote:
>Hi Everyone,
>
>I know this topic has been debated for years on the listserv with m= any helpful responses, but I thought I'd ask again to help me clarify t= hings.
>
>I have a dataset of 291 firms over 15 years (N=3D4365). What, in es= sence, is the "best" model to run for this analysis? I unders= tand the difference between fe and re and the use of the Hausman test, = which, for me, is n.s. But is there really an inherent difference betwe= en xtreg, re and regress, cluster(id)? The OLS regress, cluster(id) nor= mally gives me "better" results, but I certainly want to unde= rstand if it is "correct". I also understand that xtreg, re c= an decompose into OLS. So, if you had these data across t= his panel with a continuous DV and continous and binary IVs, what would= you run? The DV is the Cumulative Abnormal Return from a 3-day event w= indow (-1, +1)and the IVs are firm financial characteristics--ROA, volu= me, reputation dummy, etc.

*
*   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/