[R] pvclust help...
Carson Farmer
cfarmer at uvic.ca
Tue May 27 06:25:57 CEST 2008
Hello List,
I have approximately 6390 cases, each with the same three variables
(max, start, and sum). I would like to clustering the 6390 cases into
groups based on the three variables, and determine p-values for the
clusters based on bootstrap re-sampling. I can generate the clusters
using the normal hclust procedure, but it does not seem to work with
pvclust?
The pvclust help indicates that:
> For data expressed as /(n, p)/ matrix or data frame, we assume that
> the data is /n/ observations of /p/ objects, which are to be clustered.
In my case I want /p/ observations of /n /objects, so I have transposed
the data, but pvclust seems to get stuck on bootstrap sample size r=0.33.
Any help/guidance would be greatly appreciated. I am running R-2.6.0 on
Ubuntu Hardy-Heron, using pvclust-1.2-0.
Carson
More information about the R-help
mailing list