Hi R people
I am using regsubsets() to fit large numbers of models and collect summary
statistics in order to perform a Bayesian analysis of multi-way ANOVA with
specific prior information. In general the variables have differing numbers of
levels >=2. This works well but with variable of more than 2 levels there are
naturally some arbitrary decisions about which treatment contrasts to include
in the model matrix. It's obviously easy to choose the 'standard' contrasts
using
options( contrasts = c( "contr.sum", "contr.poly" ))
with contr.helmert, contr.sum or contr.treatment as required.
However, what I would really like is a set of all pairwise comparison
contrasts, and preferably in a form that can be fed straight to regsubsets() so
that all labelling, fitting, output etc is automatically dealt with.
Are there any functions or libraries in R that will allow me to do this without
too much grief?
Thanks in advance
Michael Hopkins
Algorithm and Statistical Modelling Expert
Upstream
23 Old Bond Street
London
W1S 4PZ
Mob +44 0782 578 7220
DL +44 0207 290 1326
Fax +44 0207 290 1321
hopk...@upstreamsystems.com
www.upstreamsystems.com
[[alternative HTML version deleted]]
______________________________________________
R-help@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.
