R help - Jul 2011 - biglm() and NeweyWest()

Dear all,

I am working on a large dataset and need to use biglm() to perform OLS
regressions. I have detected significant ARCH effects which I try to account
for using the Newey-West correction. 

So far, I have worked with NeweyWest() in the sandwich package. NeweyWest()
however seems to be unable to handle an object of class "biglm".

Looking into the code, I figured out that NeweyWest() calls i) bwNeweyWest()
and ii) vcovHAC(), and that both of these functions use estfun(x)... which
is unable to handle the object. I tried:

est <- biglm(y~x1+x2,...)
res <- y - predict(est,...)

estfun<-function(res,x1,x2){
return(res*cbind(1,x1,x2))
}	

...but NeweyWest() still did not work.

Is there a way to work around this problem? Do you know a package that is
able to handle "biglm" objects? 

Thanks a lot for your help!
Philipp


____________
EBS Universitaet fuer Wirtschaft und Recht i.Gr.
FARE Department
Wiesbaden/ Germany
www.ebs.edu


-----
____________________________________
EBS Universitaet fuer Wirtschaft und Recht
FARE Department
Wiesbaden/ Germany
http://www.ebs.edu/index.php?id=finacc&L=0
--
View this message in context:
http://r.789695.n4.nabble.com/biglm-and-NeweyWest-tp3692600p3692600.html
Sent from the R help mailing list archive at Nabble.com.

On Tue, Jul 26, 2011 at 12:29 AM, Philipp Grueber
<philipp.grueber at ebs.edu> wrote:
> Dear all,
>
> I am working on a large dataset and need to use biglm() to perform OLS
> regressions. I have detected significant ARCH effects which I try to
account
> for using the Newey-West correction.
>
> So far, I have worked with NeweyWest() in the sandwich package. NeweyWest()
> however seems to be unable to handle an object of class "biglm".
>
An object of class 'biglm' doesn't contain enough information to
compute the Newey-West estimator. The whole point of biglm() is that
it works with one observation at a time and doesn't store all the
data.  You'll probably need to do nearly all the work for the
Newey-West estimator yourself -- you can get the fitted values and the
information matrix out of biglm(), but not much else.

  -thomas

-- 
Thomas Lumley
Professor of Biostatistics
University of Auckland