Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

Re: st: Marginal effects for mvprobit after imputation

From   Austin Nichols <>
Subject   Re: st: Marginal effects for mvprobit after imputation
Date   Mon, 12 Jul 2010 09:41:58 -0400

Laura Cyron :
You probably have to think through what kind of marginal effect you
want to estimate.  In general, if you can predict after estimation,
you can predict at different levels and calculate average marginal
effects of various kinds (and wrap all the prediction commands in a
program to bootstrap); see also e.g.

On Mon, Jul 12, 2010 at 9:19 AM, Maarten buis <> wrote:
> --- On Mon, 12/7/10, Laura Cyron wrote:
>> I'm trying to find a way to compute marginal effects
>> after  imputation with -ice-  with Stata 11.
>> I am estimating a  trivariate probit model using the
>> -mvprobit- command
> -mvprobit- is a user written program. I am assuming you
> are using the latest version which you downloaded by
> typing -ssc install mvprobit, replace-. It is crucial
> that you tell us where you got your user written software
> from, as there are usually multiple version floating
> around in cyberspace, and the answer often depends on
> which version you are using.
> A quick look at this program suggest that in this case
> the answer is simple though probably not to your liking:
> since -mvprobit- does not use a standard -predict-
> program, none of the -mfx-, -mfx2- (also user written,
> so...), -margins- can compute marginal effects for you.

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index