Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: Data for dependent variable changed after running reoprob


From   "Zhang, Wei (IFPRI)" <W.Zhang@cgiar.org>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Data for dependent variable changed after running reoprob
Date   Tue, 6 Mar 2012 14:47:32 +0000

Dear Nick

Thanks very much for your help. I really appreciate it. It is probably a good idea to try other ways.

Wei

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: Tuesday, March 06, 2012 3:54 AM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Data for dependent variable changed after running reoprob

. findit reoprob

STB-61  sg158.1 . . . . . . . . . . .  Update to random-effects ordered probit
        (help reoprob, ghquadm if installed)  . . . . . . . .  G. R. Frechette
        5/01    p.12; STB Reprints Vol 10, pp.266--267
        correction to solve occasional abort in maximization

STB-59  sg158 . . . . . . . . . . . . . . . . .  Random-effects ordered probit
        (help reoprob, ghquadm if installed)  . . . . . . . .  G. R. Frechette
        1/01    pp.23--27; STB Reprints Vol 10, pp.261--266
        estimates a random-effects ordered probit model

shows that -reoprob- is a user-written command from 2001, as you are asked to explain. I don't recall the author being active on Statalist for some years. You may have found a bug in -reoprob-, but you'd need to interact with the author directly, I suspect.

More positively, I suspect there are now other ways to get these models, but others can give better advice on that.

Nick

On Tue, Mar 6, 2012 at 3:31 AM, Zhang, Wei (IFPRI) <W.Zhang@cgiar.org> wrote:

> I am trying to use -reoprob- for an ordered probit model for panel datasets. The model stopped when "Fitting constant-only model" and I got an error message saying "_00000Y not found", and the data for the dependent variable was changed. In my original data, I have five levels for pest pressure from 1 to 5, but now some of the "1" values are replaced with "zero". I looked at those cases where the data has been changed and I didn't see any structural difference. In addition to changing the data, the -reoprob-  command also added "group(_000006)" to label, which was previously empty. Could anyone shed some light on why the data was changed and a label was created? A previous post by J. Li in 2008 asked similar question about error message of "_00000W not found", but I don't see an answer to that particular question.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index