Dear Brian,
Have you tried the following?
zelig(formula, data, model = "logit", robust = TRUE)
There are a whle bunch of robust options, see
http://gking.harvard.edu/zelig/docs/Additional_Inputs7.html for more
details. There are some that aren't documented there as well. Take a look
at Achim Zeileis' sandwich and zoo packages for more details.
Best,
Olivia
On 7/14/06, Brian Perron <beperron(a)wustl.edu> wrote:
The Zelig package is fantastic. I am still new to R and have found the
functionality of Zelig truly impressive. Two quick questions:
1) Is there a way to invoke a sandwich estimator for clustered
observations, similar to the robcov function in Harrell's Hmisc
package? If not, could I change the standard errors manually in from a
model (e.g., logit) generated via zelig? I am guessing there is a way,
since just about anything is possible with R. (If somebody could
provide code for doing this, that would be great.)
2) Are there any plans for adding Cox regression to the available stock
of survival models?
Thanks in advance,
Brian
-
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/