On Wednesday 01 July 2009, roger koenker wrote:
> url:www.econ.uiuc.edu/~rogerRoger Koenker
> emailrkoen...@uiuc.eduDepartment of Economics
> vox: 217-333-4558University of Illinois
> fax: 217-244-6678Urbana, IL 61801
>
> On
On Wednesday 01 July 2009, roger koenker wrote:
> It's not clear to me whether you are looking for an exploratory tool
> or something more like formal inference. For the former, it seems
> that estimating a few weighted quantiles would be quite useful. at
> least
> it is rather Tukeyesque. While
It's not clear to me whether you are looking for an exploratory tool
or something more like formal inference. For the former, it seems
that estimating a few weighted quantiles would be quite useful. at
least
it is rather Tukeyesque. While I'm appealing to authorities, I can't
resist recallin
Thanks Roger. Your comments were very helpful. Unfortunately, each of
the 'groups' in this example are derived from the same set of data, two of
which were subsets-- so it is not that unlikely that the weighted medians
were the same in some cases.
This all leads back to an operation attempting
Admittedly this seemed quite peculiar but if you look at the
entrails
of the following code you will see that with the weights the first and
second levels of your x$method variable have the same (weighted) median
so the contrast that you are estimating SHOULD be zero. Perhaps
there is some
Hi,
I am trying to use quantile regression to perform weighted-comparisons of the
median across groups. This works most of the time, however I am seeing some
odd output in summary(rq()):
Call: rq(formula = sand ~ method, tau = 0.5, data = x, weights =
area_fraction)
Coefficients:
6 matches
Mail list logo
