Thanks for the detail. This is a signal to StataCorp
to consider tweaking the help to maximise clarity
on this.
Nick
[email protected]
Ada Ma
> Nick - it is puzzling to you because you know that -est store- stores
> the results as e() class variables, may be everybody knows that too
> but I did not know that until I found that -discard- discards the -est
> store- results. And I was using Stata 9.
>
> What I knew at the time:
> (1) type -ereturn list- and I'll get the list of variables
> that are e()'s;
> (2) the results saved by -est store- are nowhere to be seen when I
> typed -ereturn list-
> (3) "-discard- discards all saved results (see ereturn)"
>
> So that was how I got confused - I looked at -help discard-, from the
> description I gathered (wrongly obviously otherwise I won't be writing
> this) that things that can be pulled out by -ereturn list-, -sreturn
> list-, -return list- would be dropped, everything else would stay.
> Little that I know the -est store- stores results as hidden e()'s.
> See, that's how I faltered. It would have made more sense to me if
> the -discard- help file simply says that it discards everything except
> the data set and the matrices. Reading the Stata 8.2 help file you've
> pulled out - IMHO it's actually clearer than that of version 9's!
>
On 6/19/06, Nick Cox <[email protected]> wrote:
> > My batting average in this thread is << 1 but on this detail
> > I am still puzzled.
> >
> > In Stata 9 the help says
> >
> > discard drops all automatically loaded programs (see [U]
> 17.2 What is an
> > ado-file?); clears e(), r(), and s() saved results (see
> return); eliminates
> > information stored by the most recent estimation
> command and any other saved
> > estimation results (see ereturn); closes any open
> graphs and drops all
> > sersets (see serset); clears all class definitions and
> instances (see
> > classutil); and closes all dialogs and clears their
> remembered contents (see
> > dialog programming).
> >
> > In Stata 8.2 the help says
> >
> > discard drops all automatically loaded programs (see [U]
> 20.2 What is an ado-file?); clears e(),
> > r(), and s() saved results (see help return and [R]
> saved results); eliminates the information
> > stored by the most recent estimation command and any
> other saved estimation results (see help
> > estimates); closes any open graphs and drops all
> sersets (see help serset); clears all class
> > definitions and instances (see help classutil); and
> closes all dialogs and clears their remembered
> > contents (see help dialogs). Use discard when
> debugging ado-files.
> >
> > If you are using an earlier version, you should indicate
> that in your
> > postings.
> >
> > Nick
> > [email protected]
> >
> > Ada Ma
> >
> > > I wrote that warning because Zurab's post reminds me of
> an occassion
> > > where I had several sets of results saved using -est
> store- and then I
> > > came to a spot where Stata recommends that I try
> -discard-. I tried
> > > and then whoosh, a lot of stuff (more than I thought I'd bargained
> > > for) was gone.
> > >
> > > Most commands in Stata implement changes that are quite quickly /
> > > easily reversed, -discard- can potentially wipe out
> regresison results
> > > that take hours to run. You're of course right in saying that the
> > > help file should be read but in my case it didn't help v
> much as the
> > > -discard- help file says nothing about -est-.
> >
> > > On 6/18/06, Nick Cox <[email protected]> wrote:
> >
> > > > The warning about -discard- is also puzzling here.
> > > > The implication that it is a dangerous command
> > > > is correct, but only if you don't read the help
> > > > and consider whether it is what you want.
*
* 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/
