Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: ivtobit and limited dependent endogenous variable


From   "Brian P. Poi" <bpoi@stata.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: ivtobit and limited dependent endogenous variable
Date   Thu, 12 Jul 2007 08:09:58 -0500 (CDT)

On Wed, 11 Jul 2007, feldman wrote:

Hi,

I know that ivtobit works well with a dummy endogenous
variable but I am not sure if
Is it ok to use ivtobit when the endogenous variable is a
limited dependent variable?  More precisely suppose that for
30% of the sample the endogenous variable takes the value 0
and for the rest it has a positive value (the distribution
of zeros in the endogenous variable is different to the
distribution of zeros in the dependent variable).

Both -ivprobit- and -ivtobit- are for use when the endogenous variable is *continuous*. As Wooldridge (2002, p. 472) emphasizes, the error terms for the structural equation and the equation for the endogenous regressor are assumed to be distributed bivariate normal, which implies that the endogenous regressor should have features of a normal distribution.


-- Brian Poi
bpoi@stata.com


Wooldridge, J. M. (2002). Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press.
*
* 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