I am trying to understand what the lda function in the MASS package calculates when there are more dimensions than examples. It is my understanding that the Fisher Linear Discriminant is not applicable in this case, because the inverse of the covariance matrix cannot be calculated.
My question is how the output is calculated when the lda function is sent a dataset with more dimensions than examples. I had expected it to fail. Is a pseudoinverse (or some other technique) used internally to allow computation of the lda in this case? Thank you in advance for your assistance. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
