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: SSM and GLLAPRED

From   Nick Cox <>
Subject   Re: st: SSM and GLLAPRED
Date   Wed, 31 Oct 2012 18:41:53 +0000

The latest versions of -ssm-, -gllamm- and -gllapred- appear to be
from SSC, as you are asked to explain.

-gllapred- won't work unless e(cmd) is set as "gllamm". -ssm- sets
e(cmd) as "ssm", so it looks as if this indeed shouldn't work. Sorry,
but I have never used either command, and I can't say what would. You
may get better solutions from users of these commands.

Please note the -cmdname- convention recommended in the FAQ.


On Wed, Oct 31, 2012 at 5:31 PM, McCullough, Kathleen
<> wrote:
> I am using the SSM command for a selection model with a first-stage logit and second stage ordered probit.  I would like to create predicted values after the command.  Since this is a wrapper for the GLLAMM command, it seems that GLLAPRED should work and the documentation seems to suggest that. However, I cannot get that to run.  It keeps saying we are not following a GLLAMM command.  Any suggestions would be appreciated.

*   For searches and help try:

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