Hi:
A Google search on 'Wishart distribution' produced a lot of hits - here's
one that might be helpful to you:
http://www.math.wustl.edu/~sawyer/hmhandouts/Wishart.pdf
Its Wikipedia entry is a bit mathematical, but the opening paragraph
explains how it arises in Bayesian inference.
On Wed, Dec 1, 2010 at 8:04 AM, GMail (KU) wrote:
> Dear R users,
>
> While I was trying to learn MCMCglmm by following MCMCglmmCourseNotes.pdf,
> I ran into a few questions, and am looking for some help.
>
> First, for some reason, I cannot load the "Traffic" data set. When I did
> "data(Traffic)", I got the following warning message:
> Warning message:
> In data(Traffic) : data set 'Traffic' not found
>
There is no object named Traffic in MCMCglmm, but there is a Traffic data
frame in package MASS. Perhaps that is the one you need... BTW, I found this
through package sos:
library(sos) # install if necessary
findFn('Traffic')
and it seems to be uniquely named.
>
> Since I failed to load the Traffic data set, I was not able to following
> the examples in the MCMCglmmCourseNotes.pdf. (By the way, I am using the
> latest version of R (R version 2.12.0) on Mac OS X Snow Leopard (v. 10.6.5).
> Can any one help me solve this problem?
>
> Second, I understand that MCMCglmm uses Inverse-Wishart distribution to
> specify a prior for variances. But, unfortunately, I am not used to
> Inverse-Wishart distribution. Could someone explain how the inverse-Wishart
> distribution is related to other distributions, such as uniform, normal
> distributions, that were often used in WinBUGS/JAGS? I know how to specify
> priors using uniform and normal distributions in WinBUGS/JAGS, but could not
> figure out how to use the inverse-Wishart distribution.
>
> Any comments would be highly appreciated!
>
> Young-Jin Lee
>
HTH,
Dennis
> _______________________________________________
> R-sig-mixed-models@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
[[alternative HTML version deleted]]