Hi Casey,
The GEE model and clustered standard errors model are not exactly the same.
The GEE model allows a user to specify a correlation structure within the
clusters, whereas the clustered standard errors model does not. If the
correlation structure is correct, the GEE model produces more precise
estimates of the standard error. See Zorn (2006) in PRQ for more info on
the differences between the two models.
In the logit.gee model, you can specify the correlation structure with the
"corstr" argument. The default is independence, which is basically the same
as a logit model. So your syntax would be the same as running a regular
logit.
Patrick
On Thu, Sep 4, 2008 at 4:29 PM, Casey Klofstad <klofstad(a)gmail.com> wrote:
I want to run a simple logit with clustered standard
errors. But,
Zelig does not seem to support this. Is the logit.gee model the same
as a logit with clustered standard errors?
If so, and if I want to cluster on "X2", is this all I have to do?:
z.out <- zelig(Y ~ X1, model = "normal.gee", id = "X2", data =
mydata)
Thanks!
-c
--
Casey A. Klofstad
University of Miami
Department of Political Science
Coral Gables, FL
klofstad(a)gmail.com
http://www.as.miami.edu/personal/cklofstad/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/