JISCMail - ALLSTAT Archives

Thank you everybody for your answers.

Subject: RE: Categorical data in PCA and MDS
Date: Wed, 6 Mar 2013 17:16:13 +0000
From: [log in to unmask]
To: [log in to unmask]

Ignore that advice and the agreement on allstat. It’s a mad idea and indicates they were not listening/reading.

Codes in a nominal variable are quite arbitrary – why not 1-European 2=Asian 33=American? You should NEVER do arithmetic on such a variable. What you do to include nominal values in a model is to split each value off as a binary (or dummy) variable, usually coding positive as 1: American 0/1; European 0/1 etc. Note you can leave one category out, as 0-0-0 means not American or European or ...

If your data is mainly categorical, check if PCA is appropriate.

Allan

From: A UK-based worldwide e-mail broadcast system mailing list [mailto:[log in to unmask]] On Behalf Of Angelica Neisa
Sent: 06 March 2013 15:52
To: [log in to unmask]
Subject: Categorical data in PCA and MDS

Hello everyone,

I'm working on PCA and MDS models for my data. But one of the variables is categorical (1:American, 2:European....) . I believe that I shouldn't put this variable in my model, but a statistician friend of mine told me to standardized it (X-M/s) and include it. What do you think?

Thanks,
Angelica

You may leave the list at any time by sending the command

SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.

This email and any attachments are intended for the named recipient only. Its unauthorised use, distribution, disclosure, storage or copying is not permitted. If you have received it in error, please destroy all copies and notify the sender. In messages of a non-business nature, the views and opinions expressed are the author's own and do not necessarily reflect those of Cefas. Communications on Cefas’ computer systems may be monitored and/or recorded to secure the effective operation of the system and for other lawful purposes.