Re: st: Panel VAR: Unit root tests/Cointegration

[Michael Rose <michael.q.rose@gmail.com>]

Dear everyone,

theoretically this could work: Post the resulting matrices and then do 
the tests.

But I don't know exactly which matrix to post. Is it the residuals 
correlation matrix? If so, how do I post a matrix in stata?

Thank you very much in advance!
Michael

On 10/21/2013 09:37 PM, Michael Rose wrote:
> Dear Statalist,
>
> I'd like to conduct a Pesaran test using xtcsd.ado on a Panel VAR 
> regression using Ida Love's pvar.ado (available through the World Bank 
> website).
>
> But xtcsd doesn't like the varlist (so says the error). It seems that 
> xtcsd only works with panel regression results steeming from xtreg. 
> xtreg is Stata's build-in Unit Root Test. However they give wrong 
> results since all tests listed in xtreg assume no cointegration among 
> the panel units. Therefore the Pesaran test.
>
> But how can I run a Pesaran test (command: xtcsd, pesaran) on the 
> results of pvar? Does anyone have experience on that topic?
>
> Greetings
> Michael
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/