How to choose between fixed-effects and pooled logistic regression (xtlogit ..., fe or logit)?

[shahla ebrahimi]

I would greatly appreciate if you could let me know how to choose between fixed-effects and pooled logistic regression. Exactly saying, how to decide to use

xtlogit ..., fe

or

logit

. I know likelihood ratio test is used to chose between

xtlogit ..., re

and

logit

. Also, Hausman test is used to choose between fixed-effects and random-effects.

Thanks in advance.

[Cooper Felix]

I think it depends on the structure of your data. If you have a panel data, then using xtlogit will be appropriate.

[shahla ebrahimi]

Thanks a lot. My data set is in panel format but I need to test it.

[shahla ebrahimi]

Is it possible to do lrtest. I mean estimate logit, store it. Then, estimate xtlogit ... , fe, store it. Finally, do likelihood ratio test.

[shahla ebrahimi]

According to the

https://www.stata.com/statalist/archive/2004-11/msg00366.html

, hausman test should be done in order to choose between fixed effects and pooled logit. It refers to Greene, Econometric Analysis, 3rd edition, p. 900. However, I have access to it's 7 edition.