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: Re: st: get covariance and variance matrix after mvprobit

From   Zhi Su <>
To   statalist <>
Subject   Re: Re: st: get covariance and variance matrix after mvprobit
Date   Fri, 8 Jul 2011 10:57:56 -0400

Dear Maarten,

  Thanks for the answer. I did ask the same question two months ago
and refered to your paper. And I used _b[atrhoij:_cons] to calculate
rhoij. But using athrho to calculate rho could be tediously long when
mvprobit is involved with more than 4 equations. So I ask the question
again to see whether there is an easier way to get the variance and
covariance matrix at once. I think the e(rhoij) make things easier,
even though I do not directly get the whole matrix and still need to
use e(rhoij) to build the matrix.

Date: Thu, 7 Jul 2011 18:33:47 +0200
From: Maarten Buis <>
Subject: Re: st: get covariance and variance matrix after mvprobit

- --- On Thu, Jul 7, 2011 at 3:01 PM, Zhi Su wrote:
>>  Do anyone know how to get the covariance and variance matrix for the
>> error terms of the equation after mvprobit? For example, is it
>> contained in e() function?

- --- On Thu, Jul 7, 2011 at 3:16 PM, Maarten Buis wrote:
> The answer to this question is well documented in the helpfile of
> -mvprobit-. It gives a list of results that are returned by
> -mvprobit-, one of these is e(rhoji).

Moreover, less than two months ago you asked exactly the same question
and you received an answer:

I am somewhat annoyed by this.

- -- Maarten

- --------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
- --------------------------

Zhi Su
348 Holmes Hall
Northeastern University
360 Huntington Avenue
Boston, MA 02115
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index