On Saturday, Jul 5, 2003, at 02:33 US/Eastern, Rodrigo wrote:
I am phd student. I was running a xtgls estimation in Stata 7. After
this I
tried to test the Breusch-Pagan Statistic for cross-section
independence in
the residuals. I had a problem with that becuase my screen show me
"correlation matrix of residuals is singular not possible with test.
Thanks
for help.
The residuals from an xt estimator, arrayed in a TxN matrix, where you
have T obs on each of N panel units, are the "data" from which one
wants to calculate this matrix of correlations. If you're trying to
calculate such for N>T, it will be singular, since one cannot have more
linearly independent vectors than the rank of the matrix, which cannot
exceed T if N>T.
Kit
*
* 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/
