Chow Test

[Daniela Dias]

Hello everyone,

I'm currently working in panel data and I want to choose between a fixed effects model and a pooled ols model. To do so, I think I have to perform a Chow Test. I did some research, but I couldnt find information about the stata commands in panel data and how to analyse it.
Anyone can help me?

Thank you!

[Carlo Lazzaro]

Daniela;
welcome to the list.
I do not get why your choice is not between fixed and random effect specification:
Anyway, you can find the following threads helpful: http://www.stata.com/support/faqs/st...cs/chow-tests/; http://www.stata.com/support/faqs/st...how-statistic/

[Daniela Dias]

That was very helpful, Carlo. Thanks. I think I didnt explain the situation well. I already chose between the fixed and the random effects. After doing the Hausman Test, fixed effects seemed to be the most appropriated. But now I would like to verify the fixed effects vs polled ols model. By the way, I did some research and some people say that the F statistic that appears at the end of the estimation of fixed effects is the Chow Test. Is it correct? I can analyse only that statistic or should I really go through all the steps that your threads suggests?
Thank you for your help!

[Valeria Frigerio]

If you check at the bottom of your fixed effect regression, you should see an F-test just below the table: "F test that all u_i=0:".
That test is for the null that all coefficients for fixed effects are zero, i.e. a test between fixed vs pols. If the null is rejected, then fixed effects are significant, so you should use them. Otherwise, just go for the Pooled OLS.

[Daniela Dias]

Thank you so much Valeria!