Le 2002.05.22 16:25, "Huntsinger, Reid" a ?crit
:> This measure of "discriminatory power" is derived from linear,
not
> quadratic, discriminant analysis. Univariate QDA can discriminate between
> normal distributions with equal means but different variances.
>
> help(qda) suggests that the "means" component and the
"scaling" component
> of
> the return are most of what you're looking for.
>
> Reid Huntsinger
>
> -----Original Message-----
> From: Daniel Amor?se [mailto:Daniel.Amorese at geos.unicaen.fr]
> Sent: Wednesday, May 22, 2002 9:07 AM
> To: r-help at stat.math.ethz.ch
> Subject:
>
>
> Perhaps, I should complement my previous e-mail question:
> I am trying to select variables through a quadratic discriminant
> analysis.
> It seems to me that a robust criterion is the value of the discriminatory
> power
> (i.e. the ratio of the separation of the class means to within-class
> variance)
> How can I obtain this ratio from V & R's qda() output ?
> Thanks
> (I hope this question is relevant)
> --
> Daniel AMORESE
> Lab. M2C "Morphodynamique Continentale et C?ti?re"
> UMR CNRS 6143 Caen/Rouen
> Centre de G?omorphologie
> UCBN (Universit? de Caen Basse-Normandie)
> 14032 Caen Cedex
> FRANCE
> Ph: +33 (0)2 31 56 57 19
> Fax: +33 (0)2 31 56 57 57
> Courriel: amorese at geos.unicaen.fr
> http://www.geos.unicaen.fr/perso/da/daperso.html
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
> _._
>
>
------------------------------------------------------------------------------
> Notice: This e-mail message, together with any attachments, contains
> information of Merck & Co., Inc. (Whitehouse Station, New Jersey, USA)
> that may be confidential, proprietary copyrighted and/or legally
> privileged, and is intended solely for the use of the individual or
> entity named on this message. If you are not the intended recipient, and
> have received this message in error, please immediately return this by
> e-mail and then delete it.
>
>
=============================================================================>
>
>
Thank you.
Your answer helps me a little (I regret help(qda) and "Modern Applied
Statistics
with S-plus" provide so sparse information about qda usage).
D. Amorese
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._