At useR 2006 I mentioned that it would be nice to have a way to
specify binomial links
that involved free parameters and described some experience with a
Gosset link involving
a free degrees of freedom parameter, and a Tukey-lambda link with two
free parameters.
My implementation of this
There is no reason to add ...: you could have
binomial(link = Gosset(nu=5)).
I really don't like the idea of changing system functions like make.link,
and believe it is not necessary.
On Sun, 30 Jul 2006, roger koenker wrote:
At useR 2006 I mentioned that it would be nice to have a way to
Thanks, that works splendidly.
url:www.econ.uiuc.edu/~rogerRoger Koenker
email [EMAIL PROTECTED] Department of Economics
vox:217-333-4558University of Illinois
fax:217-244-6678Champaign, IL
NIST maintains a repository of Statistical Reference Datasets at
http://www.itl.nist.gov/div898/strd/. I have been working through the
datasets to compare R's results to their references with the hope that
if all works well, this could become a validation package.
All the linear regression
Kevin, starting with your idea of sorting first, you can get some speedups
just using R. Start by comparing the base case that Bill used:
x - runif(2e6)
i - rep(1:1e6, 2)
unix.time(res0 - unlist(lapply(split(x,i), sum)))
[1] 11.00 0.16 11.28NANA
Now, try sorting and using a loop:
Hi Bill,
After playing with this some more and adding an implementation to
handle NAs in the data vector, I have run into the problem of what to
return when the only data values for a particular bin (or level) in
the data vector were NAs and the user selected na.rm=T
1. Should it return 0
Robert Gentleman [EMAIL PROTECTED] writes:
OK, that suggests setting at the options level would solve both of your
problems and that seems like the best approach. I don't really want to
pass this around as a parameter through the maze of functions that might
actually download something if
