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: Hurlde models
From
John Ataguba <[email protected]>
To
Statalist statalist mailing <[email protected]>
Subject
st: Hurlde models
Date
Thu, 3 Jun 2010 06:47:22 +0000 (GMT)
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/