Introduction to discriminant analysis

Abstract

In Chapter 12, principal component analysis (PCA) was introduced, which can be applied when you have M observations on N variables, denoted by Y₁ to Y_N. Recall that the aim of PCA is to create linear combinations of the N variables (principal components or axes), such that the first principal component (PC) has maximum variance, the second PC, the second largest variance, etc. The first PC, denoted by Z₁, is given by

$$ Z_{i1} = c_{11} T_{i1} + c_{12} T_{i2} + ...c_{1N} Y_{iN} $$

(14.1)