期望
对离散型随机变量X,其概率分布函数(probability density function,PMF)为 P ( X ) P(X) P(X),则:
E ( X ) = μ = ∑ i = 1 n X i P ( X i ) E(X)=\mu=\sum\limits^{n}_{i=1}X_{i}P(X_{i}) E(X)=μ=i=1∑nXiP(Xi)
如果等概,就退化成我们从小就接触到的平均值 E ( X ) = ∑ i = 1 n X i n E(X)=\frac{\sum\limits^{n}_{i=1}X_{i}}{n} E(X)=ni=1∑nXi。
对连续性随机变量X,其概率密度函数(probability mass function,PDF)为 f ( x ) f(x) f(x),则:
E ( X ) = μ = ∫ − ∞ + ∞ x f ( x ) d x E(X)=\mu=\int^{+\infty}_{-\infty} xf(x)dx E(X)=μ=∫−∞+∞xf(x)dx
附带说一下,累计分布函数(cumulative distribution function,CDF)是PDF的积分形式,设其为 F ( x ) F(x) F(x),则:
F ( x ) = ∫ − ∞ x f ( x ) d x F(x)=\int ^{x}_{-\infty}f(x)dx F(x)=∫−∞xf(x)dx
方差
对离散型随机变量X,其概率分布函数为 P ( X ) P(X) P(X),则:
V a r ( X ) = D ( X ) = σ 2 = E ( ( X − E ( X ) ) 2 ) = ∑ i = 1 n ( X i − E ( X ) ) 2 P ( X i ) Var(X)=D(X)=\sigma^{2}=E((X-E(X))^{2})=\sum\limits^{n}_{i=1}(X_{i}-E(X))^{2}P(X_{i}) Var(X)=D(X)=σ2=E((X−E(X))2)=i=1∑n(Xi−E(X))2P(Xi)
= ∑ i = 1 n X i 2 P ( X i ) − E ( X ) 2 = E ( X i 2 ) − E ( X ) 2 =\sum\limits^{n}_{i=1}X_{i}^{2}P(X_{i})-E(X)^{2}=E(X_{i}^{2})-E(X)^{2}