【numpy】协方差计算

numpy中计算方差/协方差的函数 cov

公式

$$cov(X,Y)=\frac{{\sum }_{i=1}^n(X_i-\bar{X})(Y_i-\bar{Y})}{N-1}$$

X,Y为等长的两个数组

结果为一个2X2的矩阵，

$$cov(X,Y)=\left\{\begin{array}{cc}cov(X,X)& cov(X,Y)\\ cov(Y,X)& cov(Y,Y)\end{array}\right\}$$

公式可以扩展为N个数组，结果为N x N的矩阵

计算

import numpy as np

T = np.array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1])
S = np.array([2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3])
T_m = np.average(T)
S_m = np.average(S)
cov_T_T = np.sum((T-T_m)*(T-T_m))/11
cov_T_S = np.sum((T-T_m)*(S-S_m))/11
cov_S_S = np.sum((S-S_m)*(S-S_m))/11

print(cov_T_T, cov_T_S)
print(cov_T_S, cov_S_S)

X = np.vstack((T,S))

print(np.cov(X))

结果