Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: 答复: st: joint estimation of OLS and logit in seemingly unrelated regressions
From
#WANG ZITIAN# <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: 答复: st: joint estimation of OLS and logit in seemingly unrelated regressions
Date
Wed, 23 May 2012 14:04:38 +0000
Then, suppose my purpose is NOT to compare the statistical difference on the same variable across two regressions, but rather to adjust the correlated error terms across two regressions, using "seemingly unrelated method" does not make any difference than estimating each individual equation using the clustering option??
Thanks Maarten.
Zitian
-----邮件原件-----
发件人: [email protected] [mailto:[email protected]] 代表 Maarten Buis
发送时间: 23 May 2012 22:00
收件人: [email protected]
主题: Re: st: joint estimation of OLS and logit in seemingly unrelated regressions
On Wed, May 23, 2012 at 3:49 PM, #WANG ZITIAN# wrote:
> However, I find that using suest command with clustering generates almost the same standard errors comparing estimating each individual regression with clustering option. Could you explain why it is the case?
-suest- does not explicitly model any correlation between the models'
error terms, it circumvents that problem by using a variance
covariance matrix of the sandwich/robust type. So I am not surprised
by what you find.
-- Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/