Re: [R] PCA and categorical data

2009-03-06 Thread Hans Ekbrand
On Fri, Mar 06, 2009 at 09:46:17AM -, Ted Harding wrote: > On 06-Mar-09 09:25:26, Prof Brian Ripley wrote: > > You might want to look into correspondence analysis, which has several > > variants of PCA designed for categorical data. > > In particular, have a look at the results of > > RSit

Re: [R] PCA and categorical data

2009-03-06 Thread Mark Difford
Hi Galandis, dudi.mix() in package ade4 does PCA using categorical and/or quantitative variables. Ordered cats are replaced by poly(x, deg=2). Squares of categoricals can also be used. The method is a generalization by Chessel of the method of Hill and Smith. Regards, Mark. Galanidis Alexandro

Re: [R] PCA and categorical data

2009-03-06 Thread Ted Harding
On 06-Mar-09 09:25:26, Prof Brian Ripley wrote: > You might want to look into correspondence analysis, which has several > variants of PCA designed for categorical data. In particular, have a look at the results of RSiteSearch("correspondence") Ted. > On Fri, 6 Mar 2009, Galanidis Alexandros

Re: [R] PCA and categorical data

2009-03-06 Thread Prof Brian Ripley
You might want to look into correspondence analysis, which has several variants of PCA designed for categorical data. On Fri, 6 Mar 2009, Galanidis Alexandros wrote: Hi all, I' m trying to figure out if it is appropriate to do a PCA having only categorical data (not ordinal). I have only fin

[R] PCA and categorical data

2009-03-06 Thread Galanidis Alexandros
Hi all, I' m trying to figure out if it is appropriate to do a PCA having only categorical data (not ordinal). I have only find the following quote: One method to find such relationships is to select appropriate variables and to view the data using a method like Principle Components Analysis (PC