X <- matrix(c(9, 1, 5, 3, 1, 2), nrow=3, ncol=2, byrow=TRUE) S <- cov(X) S
[,1] [,2] [1,] 16 -2 [2,] -2 1
# Generalized sample variance det(S)
[1] 12
# Generalized sample variance of the standardized variables R <- cor(X) det(R)
[1] 0.75
# Total sample variance sum(diag(S))
[1] 17
X <- matrix(c(12, 17, 29, 18, 20, 38, 14, 16, 30, 20, 18, 38, 16, 19, 35), nrow=5, ncol=3, byrow=TRUE) # Sample covariance matrix S <- cov(X) S
[,1] [,2] [,3] [1,] 10 3.0 13.0 [2,] 3 2.5 5.5 [3,] 13 5.5 18.5
# Generalized sample variance det(S)
[1] 1.065814e-14
# Verify that the third column of the data matrix is the sum of the first two columns X[, 3]
[1] 29 38 30 38 35
X[, 1] + X[, 2]
[1] 29 38 30 38 35
S <- matrix(c(4, 3, 1, 3, 9, 2, 1, 2, 1), nrow=3, ncol=3, byrow=TRUE) R <- matrix(c(1, 1/2, 1/2, 1/2, 1, 2/3, 1/2, 2/3, 1), nrow=3, ncol=3, byrow=TRUE) # Generalized variances det(S)
[1] 14
det(R)
[1] 0.3888889
# Sample covariance matrix to correlation matrix cov2cor(S)
[,1] [,2] [,3] [1,] 1.0 0.5000000 0.5000000 [2,] 0.5 1.0000000 0.6666667 [3,] 0.5 0.6666667 1.0000000
We can observe that R is the corresponding correlation matrix of S.