Hi everybody, I'm performing a cluster analysis (pkg "cluster") on a dataset which includes 15 variables: is there a way to know how much each variable weighs on the final clustering output? Thanks Alessandro
alessandro carletti wrote:> Hi everybody, > I'm performing a cluster analysis (pkg "cluster") on a > dataset which includes 15 variables: is there a way to > know how much each variable weighs on the final > clustering output?What kind of cluster method? Most cluster methods are based on some distance between observations (or clusters), hence not related to weights of variables. Uwe Ligges> Thanks > > Alessandro > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Assuming you don't end up with too many clusters, you could take the classification and use it as the target for a tree, random forest, discriminant analysis or multinomial logistic regression. The random forest may be the best option.> -----Original Message----- > From: alessandro carletti [mailto:alxmilton at yahoo.it] > Sent: Friday, October 28, 2005 5:42 AM > To: rHELP > Subject: [R] clustering > > > Hi everybody, > I'm performing a cluster analysis (pkg "cluster") on a > dataset which includes 15 variables: is there a way to > know how much each variable weighs on the final > clustering output? > Thanks > > Alessandro > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html >
Hi, All: Does R have such packages for spatial epidemiology? Many thanks,
It depends on what you want to do. There are many spatial packages that can
be used for a variety of epidemiological analyses. Take a look at this link
http://sal.uiuc.edu/csiss/Rgeo// for a nice summary of the current spatial
tools in R.
If you are looking for something that will perform scan statistics along the
lines of SaTScan, the package DCluster will probably be your best option.
Follow this link
http://finzi.psych.upenn.edu/R/library/DCluster/html/00Index.html for a list
of options available in this package
Finally, try RSiteSearch("spatial epidemiology") for some interesting
threads on the subject
I hope this helps
Francisco
>From: Hongyu Sun <sun at cae.wisc.edu>
>To: rHELP <R-help at stat.math.ethz.ch>
>Subject: [R] spatial epidemiology
>Date: Tue, 08 Nov 2005 13:54:56 -0600
>
>Hi, All: Does R have such packages for spatial epidemiology? Many thanks,
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide!
>http://www.R-project.org/posting-guide.html