IJSRP, Volume 7, Issue 8, August 2017 Edition [ISSN 2250-3153]
Jude C Obi
Abstract:
This study is a follow-up to earlier publications on the relationship of least squares regression to FDA. In particular, we focus on the paper; On the Regression Discriminant Analysis (RDA), and its Identical Relationship to the Fisher’s Discriminant Analysis, and argue that since ˆb is a vector of coefficients for the least squares regression, given that y 2 (+1,1), then a substitution of ˆb with either ˆb ridge or ˆblasso gives a regression discriminant analysis variant.