Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: robust/clustered standard errors with ivprob/divprob

From   "Mark Schaffer" <>
Subject   Re: st: robust/clustered standard errors with ivprob/divprob
Date   Mon, 13 Sep 2004 16:35:48 +0100


I'll venture a very incomplete reply...

If no one has responded, I think you will probably have to take 
ivprob and change the code yourself.

And while you're at it, you might want to change the code a bit.  I 
was just looking at it, and noticed that a loop starting around line 
30 generates fitted values and residuals, but they are created as 
floats instead of doubles.  I wonder if this could generate numerical 
inaccuracy problems; doubles would be safer.


Date sent:      	Mon, 13 Sep 2004 11:14:52 -0400 (EDT)
From:           	Lisa Powell <>
Subject:        	st: robust/clustered standard errors with ivprob/divprob
Send reply to:

> Greetings,
> Given the high traffic of discussion on isues related estimation with
> endogenous variables, I though that I would repost my question from
> last week in the hopes of receiving any replies.
> I am using ivprob and divprob and would like to correct the standard
> errors based on clustering. The cluster option is not available in
> these ado files files -- does anyone have any suggestions as to how I
> could estimate a dichotmous second stage with an endogenous regressor
> that is linear (hence a linear first stage) and cluster the st errors
> in the second stage?
> Has anyone adpated the ivprob/divprob program with the cluster option?
> Any help would be much appreciated!
> Thanks,
> Lisa
> *
> *   For searches and help try:
> *
> *
> *

Prof. Mark E. Schaffer
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS  UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator

*   For searches and help try:

© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index