[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: error message r(2000)
Thank you all for your comments/suggestions. I've detected the cause of the
problem; I had failed to realise all along that some of the regressors were
(by definition) missing for the subset of observations I was examining.
There is no problem with the do file after eliminating those regressors.
From: Nick Winter <firstname.lastname@example.org>
You probably also need to tell us more about the command(s) you issues
that led to this error, if we are to give you clearer responses about why
it is happening.
At 03:51 PM 3/31/2005 +0100, you wrote:
This should not seem surprising, as it is exactly what the
-regress- message is telling you.
You may find -nmissing- useful for a concise statement
of missing values. Do a -search- to find locations.
On the whole, you need to understand what local macros
are before you can use them effectively.
> I've tried giving the -count- command again within the do
> file, using the
> actual list of variables in `varlist', separated by commas,
> and the result
> is 0. So there are actually no observations. I've got to
> figure out why
> there are no observations. Thanks a lot, Kit, Ronan, and
> Roger for your help.
* For searches and help try: