Bookmark and Share

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]

Re: st: Hurlde models


From   "Michael N. Mitchell" <[email protected]>
To   [email protected]
Subject   Re: st: Hurlde models
Date   Thu, 03 Jun 2010 00:00:33 -0700

Dear Jon

  I wonder if this article from the Stata Journal about hurdle models is helpful.

http://www.stata-journal.com/sjpdf.html?articlenum=st0040

  I hope this might help,

Michael N. Mitchell
Data Management Using Stata      - http://www.stata.com/bookstore/dmus.html
A Visual Guide to Stata Graphics - http://www.stata.com/bookstore/vgsg.html
Stata tidbit of the week         - http://www.MichaelNormanMitchell.com



On 2010-06-02 11.47 PM, John Ataguba wrote:
Dear members

This may sound simple but I have been trying to understand using the Hurdle models.  This is a two-part model but I wonder if the two parts are estimated in some sort of independent ways.  By this I mean running the probit (or any other binary choice model) in the first instance, and for those that cross the hurdle, run a truncated-at-zero count model for instance.

Basically my question is then: Do we need anything estimated from the first part introduced into the second part to correct for any thing?  This is because the Heckman model (though this is two-step and not two-part) includes something like the inverse of Mill's ratio.  I was just wondering if you would need something similar to this in the second part of a hurdle model.

Any useful suggestion on this will be appreciated.

Regards


Jon






*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index