[S] problem with multiple discriminant analysis

Stephanie Manel (Stephanie.Manel@univ-pau.fr)
Fri, 3 Apr 1998 10:05:59 +0200 (METDST)


We want to do a multiple Discriminant analysis with Splus4.0 (release 3)
using discr function to calculate the percentage of good recognition.
We apply discr function to our data : we have two categories : 0 or 1, and a
matrix of explain variables : 32 colomns and 180 rows.
We have applied discr function :
>lf.discr<-discr(lf,size)

But we don't know how to do after to calculate the percentage of good
classification?
and the exemple in the help are very short....

Thank you for any help you may be able to offer.

*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§

* Stéphanie Manel * Stéphanie.Manel@univ-pau.fr
* Département d'Ecologie
* Universite de Pau et des pays de l'Adour
* UFR Sciences et Techniques * Tél : 33(0)5/59/50/05/03
* 12 bis avenue de Mounédé * Fax : 33(0)5/59/50/05/08
* 64100 BayonneFRANCE

*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§*§

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news