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: RE: bootstrap for the simultaneous probit model
From
<[email protected]>
To
<[email protected]>
Subject
st: RE: bootstrap for the simultaneous probit model
Date
Tue, 22 May 2012 09:21:18 +0100
Assuming your model is identified (check its coherence), couldn't you
fit this model using either -cmp- or -mvprobit- (both on SSC) or using
the generalised multivariate probit code provided with Cappellari and
Jenkins's article in Stata Journal 6(2), 2006 [article is free download
from SJ site]?
No need to bootstrap in these cases.
Stephen
------------------------------
Date: Mon, 21 May 2012 08:04:55 -0700 (PDT)
From: "David Radwin" <[email protected]>
Subject: st: RE: bootstrap for the simultaneous probit model
Bilge,
This is only a partial answer, but see
http://www.stata.com/statalist/archive/2008-02/msg00394.html and also
http://sekhon.berkeley.edu/papers/FreedmanSekhon.pdf .
Freedman, David A., and Jasjeet S. Sekhon. 2010. Endogeneity in Probit
Response Models.
Political Analysis 18 (2): 138-150.
Rivers, Douglas, and Quang H. Vuong. 1988. Limited Information
Estimators
and Exogeneity
Tests for Simultaneous Probit Models. Journal of Econometrics 39:
347-366.
David
- --
David Radwin
Research Associate
MPR Associates, Inc.
2150 Shattuck Ave., Suite 800
Berkeley, CA 94704
Phone: 510-849-4942
Fax: 510-849-0794
www.mprinc.com
> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of bilge
> Sent: Sunday, May 20, 2012 10:33 AM
> To: [email protected]
> Subject: st: bootstrap for the simultaneous probit model
>
> Dear all,
> I am trying to estimate the following simultaneous probit model (model
6
> of
> chapter 8 in Maddala(1983)) :
>
> (1) Y1* = a*Y2* + b*X1 + e1
> (2) Y2* = b*Y1* + c*X2 + e2
>
> where Y1 and Y2 are two endogeneous binary variables; X1 and X2 are
two
> sets
> of exogenous variables of Y1 and Y2 respectively; e1 and e2 the error
> terms.
> Y1 and Y2 are endogenously determined by each other.
>
> As far as I know stata does not have a program for this estimation
> directly
> and that the 2-step probit estimation outlined in Maddala is
> straightforward. However, the standard errors have to be corrected. A
way
> to
> correct the variance-coveriance matrix is by bootstrapping. However, i
am
> not familiar with this method.
> Can anyone guide me in implementing this method?
> Also, do I bootstrap in both steps of the probit estimations? Do i
change
> my
> coefficients according to bootstrap results?
>
> I will appreciate your help a lot.
>
> Best,
>
> Bilge
Please access the attached hyperlink for an important electronic communications disclaimer: http://lse.ac.uk/emailDisclaimer
*
* 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/