Is there anyone out there have a program or information for calculating
the Fisher discriminant vectors?
The Fisher discriminant vectors is the eigenvectors for the following
equation:
A*x = lambda*Bx
where A is the between class scatter matrix and B is the within class scatter
matrix. The rank of B is n for large sample data set, however, the rank of A
is usually n-1 or less.
Personal e-mail is welcomed. Thank you.
Maranatha
mchung at ee.tamu.edu
