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

st: sandwich estimator for two-stage models


From   "Meike Wollni" <mw323@cornell.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: sandwich estimator for two-stage models
Date   Sat, 5 Nov 2005 17:55:19 -0500 (EST)

Dear stata-list users:
I am estimating a two-stage model: Probit in the first stage, gls in the
second.

I derive the sandwich estimate of variance to obtain corrected standard
errors for the second stage following the description in:

Hardin, J.W.: The robust variance estimator for two-stage models. The
Stata Journal (2002) 2, No. 3, pp. 253-266

My problem is that I want to include two generated regressors in the
second stage.
P = bX + cS_hat + dC_hat + e

where P-hat and C_hat are generated from two seperate probit models in the
first stage. (X is a matrix of exogenous vars, and b, c, d are unknown
vectors of parameters, and e is an error term)

Can anybody help me on how to adjust Hardin's equation for the sandwich
estimate of variance to account for a second generated regressor in the
second stage? Any suggestions will be very much appreciated.

Thank you
Meike






*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index