Erin wrote:
> I'm using the estout command after running a mulitnomial
> logit regression and then storing the estimates.
...
> If I put the variable names in the keep () portion of
> the command, I only get the coefficients from the first
> equation.
If a coefficient specified in the -keep()- option appears
in several equations, only the first occurrence of the
coefficient will be printed, unless the relevant equations
are explicitly specified in -keep()-. Thus, Erin should
type something like -keep(eq1:var eq2:var ...)-. Note that
it may be more convenient to use the -drop()- option, which
applies to *all* equations. Example:
. sysuse auto
. mlogit rep78 price weight foreign if rep78>=3
. estout, keep(weight)
.
b
4
weight .0004082
. estout, keep(4:weight 5:weight)
.
b
4
weight .0004082
5
weight -.0013343
. estout, drop(price foreign _cons)
.
b
4
weight .0004082
5
weight -.0013343
I think that it would indeed be superior if -keep()-
automatically kept the coefficients from all equations.
However, -estout- passes the -keep()- option to the
internal call of official -estimates table-. Thus,
-estimates table- would have to be fixed, not -estout-.
ben
*
* 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/
