Re: [R] 2 LDA

From: Huntsinger, Reid <reid_huntsinger_at_merck.com>
Date: Thu 16 Jun 2005 - 07:08:23 EST


Note that p1 and p2 depend only on d1 - d2. Have a look at a book like Ripley's "Pattern Recognition and Neural Networks" or Anderson's multivariate statistics book for further information.

Reid Huntsinger

-----Original Message-----
From: r-help-bounces@stat.math.ethz.ch
[mailto:r-help-bounces@stat.math.ethz.ch] On Behalf Of Steve Adams Sent: Wednesday, June 15, 2005 4:54 PM
To: r-help@stat.math.ethz.ch
Subject: [R] 2 LDA

Hi,

I am using Partek for LDA analysis. For a binary response variable, it generates 2 discriminant functions, one for each of the 2 levels of the response variable. And I can simply calculate 2 discriminant scores (say d1 and d2) for each sampples using the 2 discriminant functions, then I can use the following formula to compute the posterior probability for the sample:

p1=exp(d1-d2)/(1+exp(d1-d2))
p2=1/(1+exp(d1-d2))

In R, the lda function only generates 1 discriminant function, and exactly the same posterior probability as the LDA function in Partek does.

My question is what's the difference (or relationship) between the above 2 LDA functions, where I can find the reference for the partek LDA function?

Thanks

Steve



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Thu Jun 16 07:14:39 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:32:43 EST