[Priya Sharma]

I am still a little bit confused. How would I check for endogeneity in my model? Do I do the Ramsey Reset Test?
(but when I try to check for omitted variable bias, the command -ovtest- works only after -regress-)

[Priya Sharma]

.

[Carlo Lazzaro]

Priya:
see reply #3 at the thread I previously referred you to.

[Priya Sharma]

When the Hausman Specification Test is conducted. The null states that there is no correlation between the errors and the regressors. hence when the p-value is high, we fail to reject the null. SO there is a possibility that there is no endogeniety, so the houseman test suggest that we use the RE model right?

[Carlo Lazzaro]

Priya:
what you mentioned is not the null that -hausman- considers when comparing -fe- vs -re- specification.
Take a look at -hausman- entry in Stata .pdf manual (and related references) for more details.

[Priya Sharma]

And does the chi2 statistic determine the overall significance of the model?

[Carlo Lazzaro]

Priya:
if you refer to -xtreg,re-, you're right.