Re: st: RE: ivregress with2sls and clustered standard errors

[Charles Koss <hqtiger@gmail.com>]

Why are you clustering your standard errors by "in various schools"
and at the same time using "school fixed effects"?

-- 
Charles Koss
http://charlesonnet.blogspot.com

On Sat, Mar 26, 2011 at 4:29 PM, DE SOUZA Eric
<eric.de_souza@coleurope.eu> wrote:
> You should do what the warning suggests: use the partial option. If you are introducing a large number of dummy variables, one for each school, this could create problems.
> The partial option is explained very well in the paper by the three authors:
> http://www.stata-journal.com/article.html?article=st0030_3
> You can download it for free
>
>
> Eric de Souza
> College of Europe
> Brugge (Bruges), Belgium
> http://www.coleurope.eu
>
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Averett, Susan L
> Sent: 26 March 2011 22:20
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: ivregress with2sls and clustered standard errors
>
> Yes, I've tried that. I receive a different error message...
>
> Warning: estimated covariance matrix of moment conditions not of full rank.
>         standard errors and model tests should be interpreted with caution.
> Possible causes:
>         singleton dummy variable (dummy with one 1 and N-1 0s or vice versa) partial option may address problem.
>
>
> I am also wanting to include school fixed effects in my model as well as cluster by school. Could this be the problem?
> I do not get the message above when I drop the school fixed effects.
>
>
> ----- Original Message -----
> From: "DE SOUZA Eric" <eric.de_souza@coleurope.eu>
> To: statalist@hsphsun2.harvard.edu
> Sent: Saturday, March 26, 2011 4:41:33 PM
> Subject: st: RE: ivregress with2sls and clustered standard errors
>
> Have you tried the Stata program written by Baum, Schaffer and Stillman: ivreg2 ?
> -ssc describe ivreg2 -
>
>
> Eric de Souza
> College of Europe
> Brugge (Bruges), Belgium
> http://www.coleurope.eu
>
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Averett, Susan L
> Sent: 26 March 2011 21:08
> To: statalist@hsphsun2.harvard.edu
> Subject: st: ivregress with2sls and clustered standard errors
>
> I am trying to run a model using ivregress with 2sls as my estimator. I have a data set of individuals who are clustered in various schools. I am using school fixed effects in my model but I am trying to cluster my standard errors on the school variable AND also obtain the overidentification tests from
>
> estat overid
>
> However, when I use the cluster command, I get an error:
>
> robust tests of overidentifying restrictions after 2SLS
>  estimation not available with cluster-robust standard errors
>
> I do not get this same error if I use robust standard errors. What is the option if this error is received? Is is possible I do not need to cluster my standard errors?
>
>
> Susan A
> *
> *   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/
>
> *
> *   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/
> *
> *   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/
>
> *
> *   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/
>

*
*   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/