Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: evaluating marginal effects in trivariate probit model


From   "Stephen P. Jenkins" <stephenj@essex.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: evaluating marginal effects in trivariate probit model
Date   Fri, 13 Jul 2007 08:47:30 +0100

The following didn't show up in yesterday's Statalist Digest, so I am
re-sending it

Stephen

> -----Original Message-----
> From: Stephen P. Jenkins [mailto:stephenj@essex.ac.uk] 
> Sent: 12 July 2007 11:42
> To: 'statalist@hsphsun2.harvard.edu'
> Cc: 'Lorenzo Cappellari'
> Subject: st: evaluating marginal effects in trivariate probit model
> 
> 
> ------------------------------
> 
> Date: Wed, 11 Jul 2007 14:51:57 -0400
> From: "Xie,  Jia" <jiaxie@indiana.edu>
> Subject: st: evaluating marginal effects in trivariate probit model
> 
> Dear Sir or Madam,
>    I am using trivariate probit model, the command I am using is
> "mvprobit", however, there is problem for me to get the marginal
> effects. Can you tell me how to evaluate marginal effects after a
> trivariate probit regression? Many thanks.
>    Best
>    Jia
> ---------------------------------
> 
> There is no default marginal effects output after -mvprobit-.
> 
> There are a large number of possible definitions of ME, 
> related to whether one wants to consider effects for joint, 
> conditional or marginal probabilities. (Just look at 
> -biprobit- possibilities in the Reference Manual , and note 
> that the problem gets more complicated the more equations 
> there are.)  Note too that, for the same reason, LIMDEP 
> chooses only one of many possible definitions of ME in the MV 
> Probit context.
> 
> Note that -mvprobit-'s postestimation program, -mvppred-, can 
> produce predicted probabiliies of either all outcomes = 1, or 
> all outcomes = 0. So, one way of getting some MEs would be to 
> create some artificial observations with particular sets of 
> characteristics, then use Stata's capacity for out of 
> (estimation) sample prediction. (Consider 2 individuals with 
> all characteristics same, except for one which differs by one 
> unit, for example.)
> 
> This advice has been given before on Statalist. Please check 
> the archives before posting.
> 
> 
> Stephen (co-author with Lorenzo Cappellari of -mvprobit-)
> ---------------------------------------------------
> Professor Stephen P. Jenkins <stephenj@essex.ac.uk>
> Institute for Social and Economic Research (ISER)
> University of Essex, Colchester, CO4 3SQ, UK
> Tel: +44 (0)1206 873374. Fax: +44 (0)1206 873151
> http://www.iser.essex.ac.uk
> Survival Analysis using Stata:
> http://www.iser.essex.ac.uk/teaching/degree/stephenj/ec968/
> 
> 

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index