st: Which model to select:Fixed effect or random effect model

["Dawit Lidia" <ghibertmk@hotmail.com>]

Dear All,

I am comparing fixed and random effect models [y[i,t] = Xb + u[i] + e[i,t]] 
using Hausman specification test. The test rejected the fixed effect model. 
The result is as follow

 chi2(18) = 17.64, Prob>chi2 =      0.4795

However, when I use Breusch and Pagan Lagrangian multiplier test for random 
effects (xttest0) the result is follow

est:   Var(u) = 0
                             chi2(1) =     1.93
                         Prob > chi2 =     0.1649

It seems random effect model is not appropriate for the data set I have. In 
addition the fixed effect model gives the following test for the presence of 
individual effects:

F test that all u_i=0:     F(47, 354) =     1.45             Prob > F = 
0.0344

It says there are individual effects and thus I can't use pooled OLS.

The above results indicated that the Breusch and Pagan Lagrangian multiplier 
test for random effects rejected the random effects model but the Huasman 
support random effect model and the fixed effect model support presence of 
individual effects. Which model can I choose if my interpretation of the 
above results is correct?

Thanks for your help.

Dawit

_________________________________________________________________
MSN Hotmail http://www.hotmail.com Med markedets beste SPAM-filter. Gratis!

*
*   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/