Panel Data (Model selection)

[Sheau Shyuan Chee]

Dear expert,

I have regressed on my panel data with pooled OLS, LSDV and random effects.

The results of Hausman test showed null hypothesis cannot be rejected, which means that random effect model is better. Am i right?
However, Breusch-Pagan test showed that OLS is preferred. This is why i am not sure which model to choose. Also, i tested for heteroskedasticity and it appeared to be significant. In conclusion, the model that is appropriate for my data is pooled_OLS with robust?

Thank you so much for helping!

[Phil Bromiley]

Welcome to Statalist. You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

Pooled OLS with robust or robust clustered seems right.

[Sheau Shyuan Chee]

Hi Phil,

Thank you for your suggestions and reply!