[R] lda discriminant functions

[Marc Feldesman]

--- Leonardo Lami <lami at faunalia.it> wrote:

> Hi list,
> I'm looking about lda function.
> I'd like to know how calcolate the value of the discriminant
> functions for the 
> original datas.
> I see that in the result object "lda" there is $scaling a matrix
> which 
> transforms observations to discriminant functions, normalized so that
> within 
> groups covariance matrix is spherical.
> 
> I'd like to have the value of the discriminant function not
> normalized.
> 

The information you need to do this is already there, but you need to
understand how to manipulate your raw data in conjunction with the lda
object to calculate the result you want.  If you don't understand the
matrix math involved and the statistics necessary, you probably won't
be successful.  Suggest you read ?lda, followed by MASS, followed by
Professor Ripley's Pattern Recognition and Neural Networks, followed by
a any good book that shows you the matrix algebra for doing the
discriminant calculations.  It isn't hard to do.  If you understand the
math to get the individual scores, normalized or unnormalized, but just
do not understand the "R-way" of doing things, check back again.  

Dr. Marc R Feldesman
Professor & Chair Emeritus
Department of Anthropology
Portland State University
Portland, OR 97207

Please respond to all emails at:  feldesmanm at pdx.edu

"Some people live and die by actuarial tables"  Groundhog Day