If your panel is balanced you can use ipshin. It estimates the t-test for unit roots in heterogeneous panels.You can also use lags of the dependent variable to allow for serial correlation in the errors(Im-Pesaran-Shin 1997). However, if your panel is unbalanced you should use xtfisher. It combines the p-values from N independent unit root test (Maddala and Wu 1999).
Mansour
>>> <[email protected]> 08/23/07 5:21 PM >>>
Dear all
I'm using panel data (42 years and 9 sectors) and the panel test for units roots don't reject the hypotheses that my regressors have a unit root. Therefore, I want to test for cointegration among my variables (the dependent variable and 2 regressors). Johansen test is the test I would use for time series and, in fact, I used it for my panel data. However, I know that the specific cointegration tests for panel data (as Pedroni (1997, 1999, 2001) or Maddala and Wu (1999)) have been shown to be more powerful than the time series analog (Johansen or Augmented Engle-Granger). Do you know if there is a command in STATA to run Pedroni or Maddala and Wu tests?
Thanks in advance.
Best,
Marta.
Please access the attached hyperlink for an important electronic communications disclaimer: http://www.lse.ac.uk/collections/secretariat/legal/disclaimer.htm
*
* 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/
*
* 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/
