st: Re: Granger with fixed effect - panel data

[Kit Baum <baum@bc.edu>]

panel unit root tests make absolutely no sense with T=2. You are  
looking for the long run behavior of a time series, and you have two  
observations on the time series. No way to do that. After all, even a  
Dickey-Fuller test on a single time series implies regressing the  
difference in X on the lagged level of X, and you have precisely one  
observation for that regression.

Kit

Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On May 3, 2008, at 02:33 , statalist-digest wrote:

> I tried the levin-lin test. Unfortunately, I get an error message :
>
> . sort id year
>
> . tsset id year
>        panel variable:  id (strongly balanced)
>         time variable:  year, 2004 to 2005
>
> . levinlin csp, lags(1)
>
> Levin-Lin-Chu test for csp        Deterministics chosen: constant
> Reducing Andrews truncation
> no observations
> r(2000);
>
> . describe csp
>
>               storage  display     value
> variable name   type   format      label      variable label
> -  
> ---------------------------------------------------------------------- 
> ---------
> csp             float  %9.0g                  CSP
>
> Do you have any idea what the problem could be?
*
*   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/