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

st: RE: Clustered errors with fixed effects in Stata 8

From   "Schaffer, Mark E" <>
To   <>
Subject   st: RE: Clustered errors with fixed effects in Stata 8
Date   Thu, 6 Mar 2008 07:22:13 -0000


> -----Original Message-----
> From: 
> [] On Behalf Of 
> Steve Shewfelt
> Sent: 05 March 2008 22:36
> To:
> Subject: st: Clustered errors with fixed effects in Stata 8
> Hello,
> I'm working with a cross-sectional dataset in which observations
> (households) are clustered at the village level. I have OLS 
> and Logit models. I would like to use fixed effects and 
> robust clustered standard errors, but the options available 
> seem to allow for either of the two but not both (robust and 
> cluster are available with regress and logit, but not with 
> xtreg or xtlogit).
> Also, I am doing some IV models (ivreg and ivprob). The 
> former allows for robust clustered errors, but the latter doesn't.
> I'm using Stata 8. Does anyone know of an ado file I can 
> download that makes it possible to do what I want to do?

With respect to your FE models, -xtivreg2- runs under Stata 8 (you also
need to install -ivreg28-, the Stata-8-compatible version of -ivreg2-)
and will do OLS as well as iV models with cluster-robust SEs.  -ivreg28-
is probably also worth installing - it has many more options and
diagnostics than -ivreg-.

-ivprob- is a user written program.  Probably your only option is to
hack the code so that it accepts a -cluster- option, but a quick look at
the innards suggests to me that modifying the section that calculations
the SEs would be a non-trivial exercise.

--Mark (xtivreg2 author and ivreg2 co-author)

Prof. Mark Schaffer
Director, CERT
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS
tel +44-131-451-3494 / fax +44-131-451-3296
> Thank you in advance,
> Steve Shewfelt
> PhD candidate in Political Science
> Yale University
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2019 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index