Re: st: esttab error - too many base levels specified

[Richard Herron <richard.c.herron@gmail.com>]

My (naive) understanding of this issue is that using the -i- prefix to
generate dummies on the fly can create different base level years in
each regression.

My general solution is to generate the dummies outside of the
regression, then drop the dummies from the -esttab- output with the
-indicate()- option. If all of my year dummies have the prefix
"year_", then I would use -indicate(year FE = year_* o.year_*)-.

Although the -eststo- package is common, please provide the the source
for user-written packages. Here -esttab- is from the -eststo- package
on SSC. It helps in troubleshooting.

HTH. Please give this a try and let us know. Also, I am interested in
solutions from more sophisticated users.

On Mon, Mar 19, 2012 at 11:00, Deborah Goldschmidt <dgoldjunk@gmail.com> wrote:
> Hello,
>
> I'm running a bunch of regressions, storing the results, and trying to
> output the results using esttab. Whenever I run my regressions and include
> state or year dummies, esttab gives me the error "too many base levels
> specified" (even though the dummies aren't included in the table).  If I run
> the regressions without the state & year dummies, there's no problem. Is
> there any way to fix this, so that I can include the dummies in my analysis
> and still get the table I want?
>
> Thank you!
> debbie
>
> here is some of my code:
> eststo clear
> eststo y1Reg: reg y1 law i.state i.year, cluster(state)
> eststo y2Reg: reg y2 law i.state i.year, cluster(state)
> eststo y3Reg: reg y3 law i.state i.year, cluster(state)
>
> esttab y* using myfile.tex, replace b(a2) se ///
>            dropped(---) star(* 0.1 ** 0.05 *** 0.01) ///
>            mlabels(, none) label nogaps nonotes noobs ///
>            noconstant keep(law) ///
>            nolines delimit(&) end(\\) posthead("\addlinespace") ///
>            nonumbers  ///
>            varwidth(46) modelwidth(21) ///
>            noisily
>
>
> too many base levels specified
> r(198);
>
>
>
> *
> *   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/

*
*   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/