st: Use of psar1 and ar1 with xtpcse

[Samuel Finkelstein <s.finkelstein73@gmail.com>]

Hi,

I am in the process of estimating a model with xtpcse and I am trying
to determine if I should be using psar1 or ar1.  I used xtserial to
identify autocorrelation in my model and I also tested for the
presence of  panel-level heteroskedasticity (which is present), so I
feel that xtpcse is appropriate.  However, I cannot determine whether
I should or should not assume panel-specific autocorrelation.  I have
seen previous discussions on statalist indicating that xttest1 can be
used to determine if there is panel-specific autocorrelation, but I
have not seen any explanation beyond that it can be used.  Could you
please let me know if there is a test that allows one to determine
whether ar1 or psar1 is most appropriate (or if xttest1 can be used
for this purpose)?

Any response would be greatly appreciated.

Thank you,

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