[R] (pscl) hurdle model and sandwich: applicable?

From: Stefan Grosse <singularitaet_at_gmx.net>
Date: Wed, 26 Mar 2008 17:38:23 +0100


Dear helpeR's,

I am estimating a hurdle model with the hurdle function of the pscl package. (negative binomial for the count data and logit for the hurdle).

I am interested in robust estimates. In stata I can do this with robust cluster(group). Since R is faster by roughly factor 20 in the model estimation I would like to use R for estimations but I like to obtain also robust estimates.

I tried the sandwich package and ended up with coeftest(model,df=Inf,vcov=NeweyWest) which gives me standard errors which are very close to those of the stata function with robust cluster(group).

Now I am wondering whether this method is applicable for the hurdle models ( I am not a statistician/econometrician) and whether it is problematic if it complains with:
"Error in estfun(x)/model.matrix(x) : non-conformable arrays" - although it still gives the test resultes.

Kind regards
Stefan

-- 
Microeconomics
University of Erfurt

______________________________________________
R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Received on Wed 26 Mar 2008 - 16:59:57 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Wed 26 Mar 2008 - 17:30:24 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive