文章目录
1. 统计量
-
统计量
设 X 1 , X 2 , ⋯ , X n X_1,X_2,\cdots,X_n X1,X2,⋯,Xn是来自总体 X X X的一个样本, g ( X 1 , X 2 , ⋯ , X n ) g(X_1,X_2,\cdots,X_n) g(X1,X2,⋯,Xn)是 X 1 , X 2 , ⋯ , X n X_1,X_2,\cdots,X_n X1,X2,⋯,Xn的函数,若 g g g中不含未知参数,则称 g ( X 1 , X 2 , ⋯ , X n ) g(X_1,X_2,\cdots,X_n) g(X1,X2,⋯,Xn)是一个统计量
-
常用统计量
-
样本均值
X ‾ = 1 n ∑ i = 1 n X i \overline {X}=\frac{1}{n}\sum\limits_{i=1}^{n}X_i X=n1i=1∑nXi
-
样本方差
未修正的样本方差 : S 0 2 = 1 n ∑ i = 1 n ( X i − X ‾ ) 2 = 1 n ( ∑ i = 1 n X i 2 − n X ‾ 2 ) S_0^2=\frac{1}{n}\sum\limits_{i=1}^{n}(X_i-\overline{X})^2=\frac{1}{n}(\sum\limits_{i=1}^{n}X_i^2-n\overline{X}^2) S02=n1i=1∑n(Xi−X)2=n1(i=1∑nXi2−nX2)
修正的样本方差: S 2 = 1 n − 1 ∑ i = 1 n ( X i − X ‾ ) 2 = 1 n − 1 ( ∑ i = 1 n X i 2 − n X ‾ 2 ) S^2=\frac{1}{n-1}\sum\limits_{i=1}^{n}(X_i-\overline{X})^2=\frac{1}{n-1}(\sum\limits_{i=1}^{n}X_i^2-n\overline{X}^2) S2=n−11i=1∑n(Xi−X)2=n−11(i=1∑nXi2−nX2)
不少人会有疑问,为啥修正样本方差的分母是 n − 1 n-1 n−1,这里给予数学证明
证明
我们首先假设总体 X X X的期望和方差存在,有 E ( X ) = μ , D ( X ) = σ 2 E(X)=\mu,D(X)=\sigma^2 E(X)=μ,D(X)=σ2,此时有以下三个等式成立
E ( X ‾ ) = μ , D ( X ‾ ) = σ 2 n E(\overline{X})=\mu,D(\overline{X})=\frac{\sigma^2}{n} E(X)=μ,D(X)=nσ2,稍后会给予证明.
为了证明修正的样本方差为无偏估计,也就是我们要证明样本方差和总体方差一样,即 E S 2 = σ 2 E{S^2}=\sigma^2 ES2=σ2
证明如下
E S 2 = E [ 1 n − 1 ∑ i = 1 n ( X i − X ‾ ) 2 ] = E { 1 n − 1 ∑ i = 1 n [ ( X i − μ ) − ( X ‾ − μ ) ] 2 } = 1 n − 1 E [ ∑ i = 1 n ( X i − μ ) 2 − 2 ∑ i = 1 n ( X i − μ ) ( X ‾ − μ ) + ∑ i = 1 n ( X ‾ − μ ) 2 ] = 1 n − 1 E [ ∑ i = 1 n ( X i − μ ) 2 − 2 ( X ‾ − μ ) ∑ i = 1 n ( X i − μ ) + n ( X ‾ − μ ) 2 ] = 1 n − 1 E [ ∑ i = 1 n ( X i − μ ) 2 − 2 ( X ‾ − μ ) ( ∑ i = 1 n X i − ∑ i = 1 n μ ) + n ( X ‾ − μ ) 2 ] = 1 n − 1 E [ ∑ i = 1 n ( X i − μ ) 2 − 2 ( X ‾ − μ ) ( n X ‾ − n μ ) + n ( X ‾ − μ ) 2 ] = 1 n − 1 E [ ∑ i = 1 n ( X i − μ ) 2 − n ( X ‾ − μ ) 2 ] = 1 n − 1 [ E ∑ i = 1 n ( X i − μ ) 2 − n E ( X ‾ − μ ) 2 ] = 1 n − 1 [ ∑ i = 1 n E ( X i − μ ) 2 − n D X ‾ ] = 1 n − 1 [ ∑ i = 1 n D X i − n D X ‾ ] = 1 n − 1 ( n σ 2 − n 1 n σ 2 ) = σ 2 \begin{aligned} E{S^2} &= E\bigg[\frac{1}{n-1}\sum\limits_{i=1}^{n}(X_i-\overline{X})^2\bigg] \\&= E\bigg\{\frac{1}{n-1}\sum\limits_{i=1}^{n}\big[(X_i-\mu)-(\overline{X}-\mu)\big]^2\bigg\}\\&= \frac{1}{n-1}E\bigg[\sum\limits_{i=1}^{n}(X_i-\mu)^2-2\sum\limits_{i=1}^{n}(X_i-\mu)(\overline{X}-\mu)+\sum\limits_{i=1}^{n}(\overline{X}-\mu)^2\bigg]\\&= \frac{1}{n-1}E\bigg[\sum\limits_{i=1}^{n}(X_i-\mu)^2-2(\overline{X}-\mu)\sum\limits_{i=1}^{n}(X_i-\mu)+n(\overline{X}-\mu)^2\bigg]\\&= \frac{1}{n-1}E\bigg[\sum\limits_{i=1}^{n}(X_i-\mu)^2-2(\overline{X}-\mu)(\sum\limits_{i=1}^{n}X_i-\sum\limits_{i=1}^{n}\mu)+n(\overline{X}-\mu)^2\bigg]\\&= \frac{1}{n-1}E\bigg[\sum\limits_{i=1}^{n}(X_i-\mu)^2-2(\overline{X}-\mu)(n\overline{X}-n\mu)+n(\overline{X}-\mu)^2\bigg]\\&= \frac{1}{n-1}E\bigg[\sum\limits_{i=1}^{n}(X_i-\mu)^2-n(\overline{X}-\mu)^2\bigg]\\&= \frac{1}{n-1}\bigg[E\sum\limits_{i=1}^{n}(X_i-\mu)^2-nE(\overline{X}-\mu)^2\bigg]\\&= \frac{1}{n-1}\bigg[\sum\limits_{i=1}^{n}E(X_i-\mu)^2-nD\overline{X}\bigg]\\&= \frac{1}{n-1}\bigg[\sum\limits_{i=1}^{n}DX_i-nD\overline{X}\bigg]\\&=\frac{1}{n-1}(n\sigma^2-n\frac{1}{n}\sigma^2)\\&=\sigma^2\end{aligned} ES2=E[n−11i=1∑n(Xi−X)2]=E{n−11i=1∑n[(Xi−μ)−(X−μ)]2}=n−11E[i=1∑n(Xi−μ)2−2i=1∑n(Xi−μ)(X−μ)+i=1∑n(X−μ)2]=n−11E[i=1∑n(Xi−μ)2−2(X−μ)i=1∑n(Xi−μ)+n(X−μ)2]=n−11E[i=1∑n(Xi−μ)2−2(X−μ)(i=1∑nXi−i=1∑nμ)+n(X−μ)2]=n−11E[i=1∑n(Xi−μ)2−2(X−μ)(nX−nμ)+n(X−μ)2]=n−11E[i=1∑n(Xi−μ)2−n(X−μ)2]=n−11[Ei=1∑n(Xi−μ)2−nE(X−μ)2]=n−11[i=1∑nE(Xi−μ)2−nDX]=n−11[i=1∑nDXi−nDX]=n−11(nσ2−nn1σ2)=σ2
-
样本标准差
S = S 2 = 1 n − 1 ∑ i = 1 n ( X i − X ‾ ) 2 S=\sqrt{S^2} = \sqrt{\frac{1}{n-1}\sum\limits_{i=1}^{n}(X_i-\overline{X})^2} S=S2=n−11i=1∑n(Xi−X)2
-
样本 k k k阶原点矩
A k = 1 n ∑ i = 1 n X i k , k = 1 , 2 , ⋯ A_k=\frac{1}{n}\sum\limits_{i=1}^{n}X_i^k,\quad k=1,2,\cdots Ak=n1i=1∑nXik,k=1,2,⋯
一阶原点矩 A 1 = 1 n ∑ i = 1 n X i = X ‾ A_1=\frac{1}{n}\sum\limits_{i=1}^{n}X_i=\overline{X} A1=n1i=1∑nXi=X ,即 样本的一阶原点矩就是样本均值
-
样本 k k k阶中心距
A k = 1 n ∑ i = 1 n ( X i − X ‾ ) k , k = 2 , 3 , ⋯ A_k=\frac{1}{n}\sum\limits_{i=1}^{n}(X_i-\overline{X})^k,\quad k=2,3,\cdots Ak=n1i=1∑n(Xi−X)k,k=2,3,⋯
A 2 = 1 n ∑ i = 1 n ( X i − X ‾ ) 2 = S 0 2 = n − 1 n S 2 A_2=\frac{1}{n}\sum\limits_{i=1}^{n}(X_i-\overline{X})^2=S_0^2=\frac{n-1}{n}S^2 A2=n1i=1∑n(Xi−X)2=S02=nn−1S2 即样本2阶中心距为样本的未修正方差
-
-
关于样本均值和样本方差的性质
设总体 X X X的均值和方差存在(不管总体 X X X服从什么分布,只要它的均值和方差存在),且有 E ( X ) = μ , D ( X ) = σ 2 E(X)=\mu,D(X)=\sigma^2 E(X)=μ,D(X)=σ2, X 1 , X 2 , ⋯ , X n X_1,X_2,\cdots,X_n X1,X2,⋯,Xn是来自总体 X X X的一个样本,则有 E ( X ‾ ) = μ , D ( X ‾ ) = σ 2 n , E S 2 = σ 2 E(\overline{X})=\mu,D(\overline{X})=\frac{\sigma^2}{n},E{S^2}=\sigma^2 E(X)=μ,D(X)=nσ2,ES2=σ2
证明如下
E ( X ‾ ) = E ( 1 n ∑ i = 1 n X i ) = 1 n ∑ i = 1 n E ( X i ) = 1 n n μ = μ \begin{aligned} E(\overline{X})&=E(\frac{1}{n}\sum\limits_{i=1}^nX_i)=\frac{1}{n}\sum\limits_{i=1}^nE(X_i)=\frac{1}{n}n\mu=\mu \end{aligned} E(X)=E(n1i=1∑nXi)=n1i=1∑nE(Xi)=n1nμ=μ
D ( X ‾ ) = D ( 1 n ∑ i = 1 n X i ) = 1 n 2 ∑ i = 1 n D ( X i ) = 1 n 2 n σ 2 = σ 2 n \begin{aligned} D(\overline{X})&=D(\frac{1}{n}\sum\limits_{i=1}^nX_i)=\frac{1}{n^2}\sum\limits_{i=1}^nD(X_i)=\frac{1}{n^2}n\sigma^2=\frac{\sigma^2}{n}\end{aligned} D(X)=D(n1i=1∑nXi)=n21i=1∑nD(Xi)=n21nσ2=nσ2
E S 2 = σ 2 E{S^2}=\sigma^2 ES2=σ2在样本方差时已经证明,不在重复。