Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

SPAM (07.60): st: RE: Heckman in panel


From   "jyotsna puri" <[email protected]>
To   [email protected]
Subject   SPAM (07.60): st: RE: Heckman in panel
Date   Thu, 13 Jan 2005 19:19:48 +0000

This mail is probably spam.  The original message has been attached
so you can view it if you wish.

Content preview:  I know this question has been posed on the list before
  with responses ranging from looking up Wooldridge and GLAMM. I have
  done the former but would like to know if the following is kosher
  before I start 'learning' GLAMM. [...] 

Content analysis details:   (7.6 points, 6.0 required)
 5.4 BAYES_99               BODY: Bayesian spam probability is 99 to 100%
                            [score: 1.0000]
 1.5 FAKE_HELO_HOTMAIL      Host HELO did not match rDNS: hotmail.com
 0.7 MSGID_FROM_MTA_HEADER  Message-Id was added by a relay
-0.0 AWL                    AWL: Auto-whitelist adjustment


--- Begin Message ---
From   "jyotsna puri" <[email protected]>
To   [email protected]
Subject   st: RE: Heckman in panel
Date   Thu, 13 Jan 2005 19:19:48 +0000
I know this question has been posed on the list before
with responses ranging from looking up Wooldridge
and GLAMM. I have done the former but would like
to know if the following is kosher before
I start 'learning' GLAMM.

I have a panel dataset. I would like to estimate a
heckman model. The regular heckman command ofcourse
estimates a pooled model. I would like to correct that.

1. Estimate the selection equation via xtprobit
2. Get estimates on the mills ratio.
3. Use the mills ratio as an 'explanatory variable' in the response
equation where only the truncated dependent variable is considered,
i.e. estimate this equation for selection =1. This
is estimated via xtreg, re etc. with the hausman test to check
for specification. If necessary use Chamberlain's modification to
get consistent and efficient estimates.

Is this a correct way of thinking about it? If yes, I would ofcourse
have to account for the inter-equation correlation between the
error terms and estimate the standard errors accordingly.
I have not started to do this, so I am not sure how hard this last
step is.

Apologies if this is a silly question.



Jyotsna (Jo) Puri
[email protected]
Doctoral candidate, University of Maryland, College Park,
Maryland


*
*   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/
--- End Message ---



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