ChebyShev 定理
For any population with mean μ, variance σ and k>1, the percent of observations that lie within the interval [μ±kσ] is at least (1−k21)
【证明】对于随机变量 X,我们有
P(∣X−μ∣≤kσ)=1−P(∣X−μ∣2>k2σ2)
而考虑 ∣X−μ∣2 的期望,有
E[∣X−μ∣2]≥E[∣X−μ∣2⋅1∣X−μ∣2>k2σ2]≥k2σ2P(∣X−μ∣2>k2σ2)