Park 1967 - is there a command in STATA to perform this test?

[arunsudarsan]

Hi all,

This is my first post in this forum.

I would like to know if there is a STATA command to perform the Park 1967 test.

It is an improvement over fixed effects panel data regression with both time-fixed and cross-section fixed effects.

Sincerely,

Arun

[Stephen Jenkins]

The way to finding an answer probably involves first reading and following the advice in the Forum FAQ, especially sections 6 and 13. Hit the black bar at the top of the page, read and act please.

[Sergiy Radyakin]

Arun, questions about Park test come periodically to the list: e.g. here. Nobody bothers to explain. Perhaps you can do better.

Apparently the test is so simple, that there is still no specialized routine for it, since it can be done with standard Stata commands.

Explanation for dummies.

See here for modified Park test. This also mentions "modified Park test".

Sergiy Radyakin

[Jeff Wooldridge]

Arun seems a bit confused. I believe he wants Stata code to estimate the panel data model and corresponding generalized least squares estimator proposed by Parks (1967, JASA), which allows for cross sectional correlation and AR(1) time series errors. The Park test for heteroskedasticity is a red herring.