Wooldridge test for autocorrelation in panel data

[Elif Krkmz]

Dear all,

I am using panel probit and want to test if there is serial correlation.
I want to ask that is it suitable to use Wooldridge test for autocorrelation in panel data if your model is non-linear?

Thank you in advance.

[Carlo Lazzaro]

Elif:
although not perfectly straight to the point, you may find the following thread helpful: http://www.stata.com/statalist/archi.../msg00278.html

[Elif Krkmz]

Carlo,

Thank you in advance. I have already checked those comments but if i am not wrong it does not provide an answer to detect autocorrelation. However, it suggest that if there is autocorrelation you can fix it by using 'robust' option. But I still wonder how to detect autocorr. if your model is non-linear like probit or logit.(by the way 'robust' option is not valid for these models in Stata 11 version).

Regards,

[Carlo Lazzaro]

Elif:
I was not abe to find out an user-written programme of the sort you're after.
I do hope that Jeff Wooldridge personally will chime in this thread, giving you the most authoritative advice on his test.
Just a closing-out aside: please (as you did in the past) inform the list about the Stata release you're working with, if it is not the last one. Thanks.

[Zdavir Muhammad]

Hey idk if you already solved your problem by yourself or what.

Im new here too. But im already in love using Stata. Im using eViews, SPSS, Amos before but i found it easier using Stata because we can do lot much about it.

I encounter same problem as you with my panel data. But i found the stata based programm couldnt solve my serial correlation problem on my panel data so i search it through internet,

The link i gave you should provide a lot of information that could solve your problem

http://www.stata.com/support/faqs/st...tocorrelation/