[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: variable not found/bootstrapping
As you sense, remote debugging is not easy. But
. drop _all
you drop all observations on all variables in your dataset.
-dropping- restricted to certain observations requires
-if- or -in-.
> I'm having trouble debugging a program. The program used to
> work but now
> something's gone wrong. (It's been about a year since I
> worked with this
> program, and I think it was in Stata 7 so I've set the version in my
> program, since I now use Stata 8.2.)
> I keep getting "variable X not found" errors, even though
> when I describe
> or tabulate the variable, it's definitely there and without
> missing values.
> In my program, I do have some
> drop _all with set obs
> commands so I can make some predictions. I also have
> estimates drop _all
> at the end of the program (before the -post- command), but I
> didn't think
> these would cause the variables to be dropped, only the observations,
> right? Also, the program runs fine without bootstrapping, so
> I thought it must be related to bootstrap.
> I won't burden the list with the entire program unless that's
> necessary to
> help me debug this. Anyone have a similar problem before?
* For searches and help try: