... the Bayes theorem, as follows: 56 Introduction to Statistical Pattern Recognition -p; 0 . . . 1+p: -p; , 0 1 2 l+Pi -pi IZ; I = (1 - py . (3.14) (3.15) Therefore, the ... Cov(my,m,} =E((m, -m,)(m, -m)} (2.30) 1 N = - Cov(y,z) . Again, E{(yl -m,.)(z;-m,)} =E(yk -rn,)E(z, -mr} =O for k+L because yl and z are independent due to the independence ... unbiased and consistent estimates of M and S respec- tively. 70 Introduction to Statistical Pattern Recognition L yy I -r 0 n Fig. 3-8 The distribution of d2. The expected value...