st: XTSUR with missing observations

[Chris Parker <cparker.phd2007@london.edu>]

Hi Statalist,

I am trying to run - XTSUR - on unbalanced panel data. I let the
program run last night and it hasn't finished over the course of about
10 hours. This seems very strange for a small (462 observations, 24
groups, 2 equations with 8 or 9 variables each) regression and given
that sureg runs extremely fast (a few seconds) on the same data. I
looked at the data in the example for - XTSUR - and realized that
while the panel is unbalanced it has data at each observation for y1
and y2. In my problem I am looking at the percentage of orders that
brokers send to two different exchanges in a broker-quarter panel.
There are more than two exchanges that they could send their orders to
but I am only looking at the fully electronic exchanges. The problem
is that the first exchange opened well before the second exchange.
This means in early time periods I have data for the percent of orders
a broker sent to exchange 1 but a missing datapoint for exchange 2
since it was not open. Does anyone know if this could be causing my
problem? If so do you know of any way to get around it?

Thanks,
Chris


Chris Parker

________________________________

PhD Candidate | Management Science & Operations
London Business School | Regent's Park | London NW1 4SA | United Kingdom
Direct line +44 (0)20 7000 8816 | Email cparker.phd2007@london.edu
*
*   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/