Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: punaf error, even with version control

From   Roger Newson <>
Subject   Re: st: punaf error, even with version control
Date   Mon, 03 Oct 2011 15:11:45 +0100

Thanks to Andrew for alerting me to this catastrophe.

I have reproduced the results that Andrew describes in my up-to-the-minute Stata/MP Version 12, with and without first typing

version 11.0

although the examples work perfectly well in Stata/IC Version 11, which I also have.

Using Ben Jann's very useful package -tr-, I have traced the problem to -nlcom-. When I type, in Stata 12,

use, clear
logit low i.race i.smoke, or robust
tr: punaf, at(smoke=0) eform

I find that the failure occurs in the program -nlcom-, when it tries to execute the line of code

scalar __000007 = log(_b[_cons])

after which Stata complains (in red error letters):

[_cons] not found

Something very strange is happening here inside -nlcom-, especially as _b[_cons] appears to be negative (unsurprisingly, as it is a log baseline odds from a logit regression). If there is no advice from Statalist, and if I cannot work out what is going on myself and do a work-around, then I think I will have to report this to Stata Tech Support.

Best wishes


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Web page:
Departmental Web page:

Opinions expressed are those of the author, not of the institution.

On 03/10/2011 13:07, wrote:
Hello everyone,

I am trying to use _punaf_ in Stata 12; it worked fine in v11, but
even the examples fail in v12, giving

[_cons] not found

I have tried version control without success- I suspect the logit
estimates are simply un-findable by _punaf_ ?
Any ideas would be greatly appreciated.

*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index