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]

st: Three-stage Heckprob selection model - how to do?

From   "Pascal Stock" <>
To   <>
Subject   st: Three-stage Heckprob selection model - how to do?
Date   Thu, 13 May 2010 21:01:47 +0200

Hello STATA Users,

I am trying to estimate the following:

1.) Decision whether to use an M&A advisor (0/1)
2.) The choice of the particular bank as advisor (0/1) among 50 possible
banks if in step 1 the decision is ?use an advisor = 1?
3.) After the particular choice of a bank as advisor in step 2 the
probability of a follow-on-acquisition (0/1).

Hence I have two steps of selection that depend on each other. How to do
this with the heckprob command? Or ist the triprobit command better suited?
Also ist it possible to do this ?by hand? by calculating the inverse mills
ratios by hand and inserting them as additional regressors? I know that the
variance of the estimated regressors has to be adjusted as the inverse mills
ratio is estimated as well and not known (Greene (2003), p. 785 and Heckman

Also does an ?oheckprob? command exist similarly to the ?oheckman? command
of an ordered probit regression as selection equation and a structural
probit equation in the second step?
An suggestions?

Kind regards

Pascal Stock, M.Sc. M.A. 
Ph. D. Candidate in Finance
Center for Doctoral Studies in Business
Graduate School of Economic and Social Sciences
& Chair of Corporate Finance
University of Mannheim

Phone: +49 (0) 621 - 181 3774
Mobile phone: +49 (0) 178 - 4981887
Internet: & 

Schloss, Mittelbau, Room M114
68131 Mannheim

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index