st: estout error: too many base levels specified

[Alexandru Cojocaru <scojocaru@gmail.com>]

Dear All,

I am estimating individual-level regressions that include a set of
country dummy variables. I would like to estimate separate models over
several subsets of countries. When I try to output the results using
Ben Jann's estout,  it complains about "too many base levels
specified", which is true given non-overlapping groups of countries,
even though I tell estout not to report the estimates on the country
dummies.

Here's a contrived example that replicates the error:

******************************************************
sysuse auto, clear
reg price mpg weight i.rep78 if inrange(rep78,1,3)
    est store m1
reg price mpg weight i.rep78 if inrange(rep78,4,5)
    est store m2
estout m1 m2, keep(mpg weight))
********************************************************
leads to:

too many base levels specified
r(198);


Is there a way around this?

Cheers,
Sandu

*
*   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/