[R] Homogenous groups building - Randomisation
spencer.graves at pdf.com
Thu Jan 19 05:37:33 CET 2006
I'm sorry, but I still do not understand, but I will make a guess:
You have data on n characteristics from each of many patients, and you
want to form N groups, with roughly k patients per group from the next
roughly k*N patients who walk into your clinic.
If this is your problem, and if you have a "training set" of data you
can use to define group selection criteria, I suggest you try various
versions of cluster analysis, including "hclust" in the "stats" package
and other functions described with "See also" in the "hclust"
If you don't have this and would still like help from this listserve,
please think very carefully about what it would take for someone to
produce a useful response to your question. Je voudrais vous aidez,
mais ce demand plus que j'ai.
Hope this helps,
a.menicacci at fr.fournierpharma.com wrote:
> Dear R-users,
> We expect to form N homogeneous groups of n features from an
> experimentation including N*n data.
> The aim is to prevent group effects.
> How to do that with R functionalitites ? Does anyone know any way enabling
> this ?
> Example :
> 100 patients are observed. 3 biochemical parameters are mesured for each
> one (Red and white globules ans glycemia).
> Patient RG RW Gly
> 1 3.4 1.38 1.62
> 2 1.8 1.19 1.55
> 3 1.9 1.26 1.77
> 4 3.0 1.29 1.72
> 5 1.9 1.09 1.72
> 6 3.3 1.31 1.63
> ... ... ... ...
> These are freakish data.
> How to compute 10 "equivalent" groups ?
> For example with closed parameters means between groups :
> Group RGmean RWmean Glymean
> 1 1.5 1.22 1.68
> 2 1.3 1.29 1.75
> 3 1.6 1.25 1.63
> 4 1.2 1.23 1.70
> ... ... ... ...
> Best regards.
> Alexandre MENICACCI
> Bioinformatics - FOURNIER PHARMA
> 50, rue de Dijon - 21121 Daix - FRANCE
> a.menicacci at fr.fournierpharma.com
> tél : 03.80.44.76.17
> R-help at stat.math.ethz.ch mailing list
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
More information about the R-help