[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: -outreg-: some general comments
Let me respond to this message by answering an unanswered
question that I posed a week or so ago-- if anyone else runs
across the same problem maybe they will find the answer
I have a 24 country survey dataset, and my outreg was not
working when I ran ordered probits and (a) included country
dummies (b) clustered by country. It was the combination of (a)
and (b) that caused the problem. The problem was successfully
diagnosed by someone not on this list -- outreg broke down
because when I ran these regressions e(df_m) was set by stata
equal to zero. I now possess a version of outreg that can cope with this
problem (at the cost of including the cut values in the table); and if
anyone runs into this problem they may obtain said ado file from me.
>-outreg- has many devoted users, and it
>also attracts a fair number of questions,
>some of which can evidently be answered by
>those who also use it, although many appear
>to get no response.
>In asking questions about -outreg-
>the following points should please be borne in mind,
>subject to correction as always.
>They may help to explain why those questions
>often receive no answers.
>1. The author of the program is no longer
>a member of Statalist, nor does he
>appear to be active with Stata. Only
>he, I suspect, could answer many of the
>questions which are posed.
>2. -outreg- is a long and complicated
>program, which makes it less likely that even
>Stata programmers can see at a glance
>what might be the answer to a question.
>That of course doesn't rule out careful
>reading of the help.
>3. -outreg- appears to have been last revised
>two years ago. That doesn't make it
>obsolete in any sense, but it may mean that
>as Stata changes it will slowly but surely be
>less able to fulfil the expectations that
>many users want to put upon it.
>* For searches and help try:
Kevin H. O'Rourke
Professor of Economics
Department of Economics and IIIS
Tel. 353 1 608 3594
Fax. 353 1 677 2503
* For searches and help try: