Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: ivregress with censored endogenous regressor

From   "Facchini Palma, Gabriel Alejandro" <>
To   "" <>
Subject   st: ivregress with censored endogenous regressor
Date   Thu, 25 Jul 2013 17:19:31 +0000

Dear listers, 

I have a question regarding the best way to do a regression when I have to instrument an independent variable that is actually endogenous and censored (obs values <0 coded as zero). 

The problem here is that the normal ivregress command would perform an OLS regression in the first stage and that's biased. 

Then I can do a 2sls manually, running first a Tobit model on my censored variable with the instrument Z, and then plugging  the predicted xhat into an OLS regression with my final continuous outcome variable Y. But then would the mechanism used to correct the se in a standard 2sls work, or there's something else to take into account in this case?

I've seen some posts where it has been recommended to use only uncensored observations. The problem with that is that the assumptions behind that being consistent are quite strong and there are efficiency losses.

In case it helps, in my case the Y var is weight at birth, the endogenous censored var is # of cigarettes the mother consumed daily during pregnancy, and the Z instrument is cigarette excise taxes changes.

thanks in advance, 


The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination, distribution, forwarding, or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited without the express permission of the sender. If you received this communication in error, please contact the sender and delete the material from any computer.

*   For searches and help try:

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