Re: [R] calculating dissimilarities in R

From: Martin Maechler <maechler_at_stat.math.ethz.ch>
Date: Tue 26 Sep 2006 - 07:55:50 GMT

Hi Elvina,

>>>>> "Elvina" == Elvina Payet <virgin@seychelles.sc> >>>>> on Tue, 26 Sep 2006 05:48:01 GMT writes:

    Elvina> ,A (BDear All,
    Elvina> I’ve got a statistical question on calculating
    Elvina> dissimilarities in R.
    Elvina> I want to calculate the different types of dissimilarities
    Elvina> on the ‘flower’ dataset found in the package
    Elvina> ‘cluster’. Flower is a data frame with 18 observations
    Elvina> on 8 variables. Variable 1 and 2 are binary, variable 3 is
    Elvina> asymmetric binary, variable 4 is nominal, variable 5 and 6
    Elvina> are ordered and variable 7 and 8 are interval scaled.

    Elvina> Commands to load the dataset in R.

> library(cluster)
> data(flower)

or data(flower, package = "cluster")

    Elvina> What are the different types of dissimilarities that can be
    Elvina> calculated on such a dataset?  
    Elvina> Do I need to group the types of variables first i.e. all
    Elvina> binary together then run the calculation?  Do I use
    Elvina> dissimilarity indices such as Jaccard or should it be
    Elvina> classification function such as ‘daisy’ which should be
    Elvina> used? 


Yes, you should use daisy() to calculate dissimilarities, particularly when you are interested in the difference between symmetric and asymmetric binary.

Do read help(daisy) and look at its examples.

Maybe this will answer all your questions or then it will help you to ask a much more specific question as suggested by the posting guide (see link below!)

      [.........]

    virgin> ______________________________________________

      [.........]

    virgin> PLEASE do read the posting guide
    virgin> http://www.R-project.org/posting-guide.html 
	    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

    virgin> and provide commented, minimal, self-contained, reproducible code.

Regards,
Martin Maechler, ETH Zurich



R-help@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 and provide commented, minimal, self-contained, reproducible code. Received on Tue Sep 26 17:58:23 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Tue 26 Sep 2006 - 08:30:08 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.