st: xtcsd, pesaran quitting because of not enough common observations across panel

[Christian Wagener <wagener.christian@googlemail.com>]

Dear Stata-Users,

using Stata 11, I am running a fixed-effect-regression on a highly
unbalanced panel data set with e.g. the following commands:

xtreg servr logempl, fe robust

After that, I would like to test for cross-sectional dependence using
the "xtcsd, pesaran" command (obtained from -ssc desc xtcsd-).
Although Pesaran's statistic seems to be able "to handle balanced and
unbalanced panels" (xtcsd help), my data seems to be too unbalanced to
work with it, since I am getting the error message:
"Error: The panel is highly unbalanced.
Not enough common observations across panel to perform Pesaran's test.
insufficient observations
r(2001);"

The dataset contains 1005 companies x 21 years of possible data =
21,105 rows. Due to firms starting up later than the starting year or
going bankrupt, the panel is highly unbalanced and the number of
observations used for the regression drops to 12206.

Are there any workarounds to test for cross-sectional dependence in
such a setting or does that question the use of this dataset in
general?

Please excuse if I have forgotten any necessary information.

Thank you for your advice, best regards

Christian
*
*   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/