[R] lda() - again.

marzban marzban at caps.ou.edu
Wed Jul 14 00:56:47 CEST 2004


I asked a question about lda() and got some answers. However, one 
question remains (which is not independent of the earlier ones):

What output does lda() produce which I can use to compute the
posteriors? I know predict(lda())$posterior will give me precisely the
posteriors, but suppose I'd like to compute them myself, outside
of R. 

So far, I have not been able to use "coefficients of linear 
discrimiants" to do this, for they don't seem to be the alpha and beta
in     log(post) ~ alpha x + beta  (this eqn being a caricature of 
LDA in Ripley).


More information about the R-help mailing list