## Re: st: Fixed Effect Estimation Results

consider a model

y[i,t] = X[i,t]*b + u[i] + v[i,t]

where u[i] are individual effects constant over time (note there is not t subscript), and v[i,t] is the "traditional" error term.

in a fixed effect model, the time-invariant u[i] are assumed to be correlated with X[i,t]. e.g. in evaluating firm performance, it is, as a rule, impossible to measure managerial skills or organizational culture so these unobservable effects will be correlated with the regressors. stata reports this correlation as corr(u_i, Xb).

rafal

Quoting "Sam Rawlings, Social Sciences and Law 03" <sr3209@bristol.ac.uk>:

> Hi There, > > I am an undergraduate student in my final year and I am quite new to > Stata, and indeed, to a lot of the techniques and models that can be > used in Stata. So my apologies if this question may seem quite simple > but I can't seem to find an explanation anywhere. > > I have estimated (several) sub-samples for a fixed effects model > using panel data. However, in order to interpret my results, I'm > slightly confused about one of the statistics given - corr(u_i, Xb). > Is there any chance anyone could enlighten me, as I can only guess > what it means and would rather know for sure when writing up my > results. I've included one of my regression results below for > reference to what I'm asking about. Any comments would be appreciated! > > Samantha > > > Fixed-effects (within) regression Number of obs = 560 > Group variable (i): country Number of groups = 112 > > R-sq: within = 0.6385 Obs per group: min = 5 > between = 0.9909 avg = 5.0 > overall = 0.9694 max = 5 > > F(3,445) = 62.01 > corr(u_i, Xb) = 0.9249 Prob > F = 0.0000 > > --------------------------------------------------------------------------- > --- > lngdpwaa | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+------------------------------------------------------------- > --- > laglngdpwaa | .6389257 .0263356 24.26 0.000 .5871681 .6906832 > lns | .1294206 .0248952 5.20 0.000 .0804939 .1783473 > lnngdelta | -.0818537 .0922956 -0.89 0.376 -.2632431 .0995357 > _cons | 3.080864 .3074474 10.02 0.000 2.476635 3.685093 > -------------+------------------------------------------------------------- > --- > sigma_u | .29921031 > sigma_e | .13863343 > rho | .82326502 (fraction of variance due to u_i) > --------------------------------------------------------------------------- > --- > F test that all u_i=0: F(111, 445) = 2.78 Prob > F > = 0.0000 > > ---------------------- > SB Rawlings, Social Sciences and Law 03 > sr3209@bristol.ac.uk > * > * 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/ >

================ Rafal Raciborski Graduate student Department of Political Science Emory University 301 Tarbutton Hall 1555 Dickey Drive Atlanta, GA 30322 404-378-9826 (home) rafal.raciborski@emory.edu http://www.roofoos.net/

* * 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/

Tag: