Another point is that your local macros
must be in the same space as that in which your
-xi: xtreg- command is issued. One of the earliest
lessons in Stata programming is that -local- macros
are precisely named. But I've occasionally
been bitten by this in the following form:
I'm defining local macros in code
in the do-file editor and doing something
else in the Command window.
Then I forget momentarily that the local macros are
invisible to the main session.
The following experiment makes the point.
Type
local foo "foo"
in the Do-file editor and then run the code.
This command will be echoed to the Results
window but
. di "`foo'"
will just show an empty string [a new line,
strictly] while
. mac li
will show no -foo-. That is because the code
in the do-file editor counts as a separate
program and what you just defined exists
only in its own space.
This isn't a major problem, but it's a small
test of understanding.
P.S. for those who know it, -c_local- is
a work-around. If you don't know it, you
shouldn't want to ask.
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Nick Cox
> Sent: 18 October 2005 18:39
> To: [email protected]
> Subject: st: RE: obvious error?
>
>
> The best test here that you and Stata
> are speaking the same language is to
>
> . mac li
>
> to see what Stata is holding in local
> macros. If you typed, as stated here,
>
> . local x1lsr age age2 hgc
>
> then the local macro -x1lsr- will
> hold "age age2 hgc" and it is entirely
> possible that the local macro -x1-
> is not defined.
>
> But I can't see what is wrong with
> your other example.
>
> Stata is not _asking_ for last estimates.
> It can't see any variable names, so is
> presuming that you are asking for a replay;
> but it can't oblige, as it cannot find
> the last estimates from that command.
>
> Nick
> [email protected]
>
> Scott Cunningham
>
> > This is stumping me. Here's a sample of my code and the error I'm
> > getting. I know I'm doing some obviously wrong, but it's
> so obvious
> > that I can't see it.
> >
> > . local y rp
> > . local x1lsr age age2 hgc
> > . local x3 lsr age age2 hgc i.year state*
> >
> > . xi: xtreg `y' `x1', fe i(id) robust
> > last test not found
> > r(302);
> >
> > I get the same error even when I run the following regression:
> >
> > . xi: xtreg `y' `x3', fe i(id) robust
> > last estimates not found
> > r(301);
> >
> > So I know it's not just the fact that I'm using the -xi- operator
> > without including the year dummies for the first FE
> regression. Why
> > is it asking me for "last estimates"?
> >
>
> *
> * 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/
>
*
* 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/
