st: How to choose between psar1 and ar(1) after xtpcse

[Tricia D <tricia201401@gmail.com>]

Hi Statalisters,


I am running a panel model with the xtpcse command in Stata for an

unbalanced panel (and the pairwise option).


Is there a test for panel-specific autocorrelation that I can run to

determine whether psar(1) an AR(1) is more appropriate?


A prior posting suggests the following syntax, which, unfortunately,

doesn't work in may case:


If you regress each cross-section i using,

forvalues i=1/108 {

display "Regress pstrmon for CUSIP" `i'

quietly ivreg pstrmon price mat age coup pstrmonprev pstrprev intr ivol

compl (precmon = precmonprev) if cusip == `i'

estimates store model `i'

more

}



Any advice is highly appreciated!!



Thanks,

Tricia
*
*   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/