Claude,
levinlin is more appropriate with small T panels and large N, that is your case.
See madfuller help and levinlin help!
Best,
Joao Ricardo Lima
2008/5/2 Claude Francoeur <[email protected]>:
> Hello,
>
> I have a balanced sample of data over 2 years. I am trying to do a granger
> causality test. I would like
>
> 1- to perform a Augmented Dickey-Fuller test for covariance stationarity
>
> 2- Control for fixed effects
>
> As vargranger does not work with panel data, I have generated a lag variable
> manually.
>
> This test works fine :
>
> regress roa roa_1 csp_1 if year==2005
>
> But, how would I go about controlling for fixed effects with such a design?
>
>
> Using the tsset command, the modified ADF test does not work :
>
> . sort id year
> . tsset id year
> panel variable: id (strongly balanced)
> time variable: year, 2004 to 2005
>
> . madfuller csp, lags(1)
> T must exceed N for sureg
> invalid syntax
> r(198);
>
> How can I perform this test otherwise?
>
> Any help would be appreciated.
>
> Thank you,
>
> Claude
>
> *
> * 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/
>
--
-------------------------------
Joao Ricardo Lima
Professor
UFPB-CCA-DCFS
+553138923914
-------------------------------
*
* 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/
