Discriminant analysis to select best treatment

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Discriminant analysis to select best treatment

Felipe Martínez-Pastor
I am designing an experiment to trial several analytic techniques on
samples submitted to different treatments. It has occurred to me that I
may use discriminant analysis to find out which kind of analysis best
reveals differences between treatments.

I have found the lda {MASS} in R. However, I am not sure if it is
adequate to my case, since it performs linear discriminant analysis but
--apparently-- not variable selection. In SAS I would use the STEPDISC
procedure, whose description is:
"uses forward selection, backward elimination, or stepwise selection to
try to find a subset of quantitative variables that best reveals
differences among the classes."

I wonder if I could use lda or other function in that way. Maybe you
could point me to any resource (I am really new in discriminant analysis).

Thank you.

--------------------oOo--------------------
Felipe Martínez Pastor, Ph. D.
Ciencia y Tecnología Agroforestal
ETSIA-UCLM
Av. España s/n
02071-Albacete (Spain)
Phone: +34 967 599 200+2581
Fax:   +34 967 599 238+2081
Mobile: +34 687 365 362
e-mail: [hidden email]
Jabber: [hidden email]



--
Ningún investigador sin contrato.
Employment rights for Spanish junior researchers.
        http://www.precarios.org

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: Discriminant analysis to select best treatment

Felipe Martínez-Pastor
Hi.
I have found stepclass {klaR}, which is similar to what I was looking
for. However, I take note of your advice en stepwise variable selection.
Greetings.

Felipe

Alexandre Santos Aguiar wrote:
> Em Dom 12 Fev 2006 16:10, Felipe Martínez-Pastor escreveu:
>> --apparently-- not variable selection.
>
> About automatic variable selection one should read this:
> http://www.pitt.edu/~wpilib/statfaq/regrfaq.html
>
>

--
Ningún investigador sin contrato.
Employment rights for Spanish junior researchers.
        http://www.precarios.org

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html