2. Principal component analysis (PCA)

2.1 Objectives

The purpose of principal component analysis is to study data resulting from the observation of p quantitative variables on n individuals, arranged in a matrix X (n x p). The objectives are :

• the "optimal" graphical representation of the individuals (lines), minimizing the distortions of the point cloud, in a subspace E _q of dimension q (q < p) of the vector space ${ℝ}^p$ ;

• the graphical representation of variables in a subspace F _q of the vector space...