Statalist The Stata Listserver


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

RE: st: marginal effects after mvprobit


From   "stata_user stata_user" <[email protected]>
To   [email protected]
Subject   RE: st: marginal effects after mvprobit
Date   Mon, 02 Apr 2007 22:19:55 +0100

Hi,

I have used the mvnp from Cappellari & Jenkins to obtain the probabilities after mvprobit, following Stepehn's suggestion.

I am looking to find a way to obtain the standard erros in order to see if the marginal effects are significant or not.

I have seen some papers which used the Cappellari & Jenkins approach, they report tables with the marginal effects and standard errors, but there is no explanation how they obtain the standard errors.

Thanks a lot for any suggestion,

Sami


From: "Stephen P. Jenkins" <[email protected]>
Reply-To: [email protected]
To: <[email protected]>
Subject: st: marginal effects after mvprobit
Date: Mon, 19 Mar 2007 23:57:05 -0000

> Date: Sat, 17 Mar 2007 19:11:36 +0000
> From: "stata_user stata_user" <[email protected]>
> Subject: st: marginal effects after mvprobit
>
> Hi all,
>
> I have a question about how to get marginal effects after mvprobit,
> I have three binary outcomes, I run the mvprobit , and I
> would like to
> obtain the marginal effects for p111,p110 .... in order to
> interpret my
> outputs.
>
>
> Thanks a lot for any help,
>
> Sami
>

This is a non-trivial exercise (just look at the relevant section of
the -biprobit- entry in the Manuals)

What you could do is derive a predicted probability for whichever
outcome combination you were interested in, given a particular set of
covariates. Then recalculate the predicted probability with a marginal
change in one covariate holding all the others at the value used in
the first calculation. This gives a change in predicted probability.
-mvppred- after -mvprobit- gives you the ability to calculate Prob
that each binary outcome = 1, and Prob that each binary outcome = 0.
(If you want other outcome combinations, then you would need to use
_gmvnp(): see Cappellari & Jenkins article in Stata Journal 6(2) --
preprint at ISER website under working papers.)

Stephen (co-author of -mvprobit-)
----------------------
Professor Stephen P. Jenkins <[email protected]>
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/index.php




*
*   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/
_________________________________________________________________
Get Hotmail, News, Sport and Entertainment from MSN on your mobile. http://www.msn.txt4content.com/

*
* 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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index