Hausman test - gives p value as dot, not even 0.00

[Nishat Alam]

Please advice why Prob>chi2 is given as a . in my test, rather than a value for me to interpret.

[Carlo Lazzaro]

Nishat:
are you sure that you're not comparing the same specification (ie, -fe- vs -fe- or -re- vs -re-)?

[Nishat Alam]

I do not believe so, but STATA is a program I have been struggling with, judging from my do file, is that a possibility?

[William Lisowski]

You must do the estimates store following the xtreg whose results you want stored, before those results have been replaced by the results of a subsequent regression,.

So your estimates store commands first saved the estimates from the immediately preceding (random effects) xtreg ..., re command with the name fixed, and then saved them again with the name random.

This is consistent with the fact that the hausman output reports identical coefficients for both regressions, as Carlo suggested.

[Nishat Alam]

Thank you Carlos and William! This is the updated do file and Hausman test, looks a lot better now

Null: random effect model is appropriate
Alternative: fixed effect model is appropriate

I'm rejecting the Null hypothesis based on the p-value being so small, and less than 0.0005




Does the result of the -hausman-test at 24.98 mean that the coefficients shared by -fe- and -re- specification are not that similar than because it is kind of a big number ?
Appreciate the help, hopefully my inferences and understanding will be okay too

[William Lisowski]

The output tells you that the of 24.98 is a chi-square statistic with 5 degrees of freedom, and thus you can reject Ho with a p-value of .0001.

Let me add some advice to improve future posts.

Take a few moments to review the Statalist FAQ linked to from the top of the page, as well as from the Advice on Posting link on the page you used to create your post. Note especially sections 9-12 on how to best pose your question. It's particularly helpful to copy commands and output from your Stata Results window and paste them into your Statalist post using code delimiters [CODE] and [/CODE], and to use the dataex command to provide sample data, as described in section 12 of the FAQ.

In particular, screen shots are generally not considered helpful except for the output of graphics commands.