Dear Chris,
To get predicted probabilities from an ordered probit, use the simqi
command. The "sim" means simulate and the "qi" is for quantities of
interest.
sumqi does something slightly different: it summarizes simulations that
have already been stored as variables in your dataset. It works a bit
like Stata's summarize command, and it does not allow the pr option. If
you remove "pr" from the command you typed, and simply enter sumqi b1 b2
b3 b4 b5, you will get means and confidence intervals for the simulated
parameters that are stored in b1, b2, b3, b4, and b5.
I hope this helps. Thanks for your interest in Clarify.
Mike Tomz
Stanford University
Chris O'Keefe wrote on 9/17/2007 5:59 AM:
Hi all-
I'm trying to use clarify to produce predicted probabilities after
estimating an ordered probit. However, it seems that Clarify 2.1 no
longer allows the option of estimating predicted probabilities using
the sumqi command. I've been entering
sumqi b1 b2 b3 b4 b5, pr
only to be told that that option is not allowed.
So, how does one compute predicted probabilities using 2.1?
Thanks,
Chris O'Keefe
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=clarify
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=clarify