Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[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)" <[email protected]>
To
"[email protected]" <[email protected]>
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: [email protected] [mailto:[email protected]] On Behalf Of Nick Cox
Sent: Tuesday, March 06, 2012 3:54 AM
To: [email protected]
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) <[email protected]> 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/