D Jimenez Rubio
> Im running a panel data with annual dummies. I would like
> to check for
> heteroskedasticity in the errors of my model. Given that in
> STATA 7 I
> havent found any specific test for hetero in fixed-random effects
> models, I have tried to run LSDV with year dummies and my provincial
> (unit) dummies and then test for het using the command whitetst. But
> when I do so, I obtain the following:
>
> matsize too small; type -help matsize-
>
> So then I type -help matsize-, and I obtain:
>
> varlist not allowed
>
> Is there anyone who can tell me why this happens? Thanks a lot!
I can't advise on the econometrics here, but,
working backwards, when I type
help matsize
I get help on -matsize-, so I can't reproduce that problem
in Stata 7 (executable of 11 June 2002). And -whitetst- is
telling you that the matsize (matrix size) is too small,
so you need to increase that, unless, that is, your problem
requires a matrix bigger than your Stata can allow (which
could be 800 * 800, unless you have Stata 7/SE).
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
