Sample Variance

最新推荐文章于 2020-11-18 15:09:45 发布

znculee

最新推荐文章于 2020-11-18 15:09:45 发布

阅读量826

点赞数

分类专栏： Mathmatics 文章标签： statistics

本文链接：https://blog.csdn.net/znculee/article/details/50372493

版权

Mathmatics 专栏收录该内容

2 篇文章 0 订阅

订阅专栏

Var (X n) = S 2 n = \sum n i = 1 ( X i - X ¯ n ) 2 n - 1

$\text{Var}\left(X_n\right)=S_n^2=\frac{\sum _{i=1}^{n}\left(X_i-\bar{X}_n\right)^2}{n-1}$

E (S 2 n) = E ⎛ ⎝ ⎜ ⎜ \sum n i = 1 ( X i - X ¯ n ) 2 n - 1 ⎞ ⎠ ⎟ ⎟

$\text{E}\left(S_n^2\right)=\text{E}\left(\frac{\sum _{i=1}^{n}\left(X_i-\bar{X}_n\right)^2}{n-1}\right)$

(n - 1) E (S 2 n) n - 1 n E (S n) 2 = = = = = = \Leftrightarrow = E (\sum i = 1 n (X 1 - X ¯ n) 2) E (\sum i = 1 n X 2 i) - E (\sum i = 1 n 2 X i X ¯ n) + E (\sum i = 1 n X ¯ 2 n) E (\sum i = 1 n X 2 i) - E (2 X ¯ n \sum i = 1 n X i) + E (X ¯ 2 n \sum i = 1 n 1) E (\sum i = 1 n X 2 i) - E (2 n X ¯ n) 2 + E (n X ¯ 2 n) E (\sum i = 1 n X 2 i) - n E (X ¯ 2 n) n E (X 2 i) - n E (X ¯ 2 n) E (X 2 i) - E (X ¯ 2 n)

$\begin{array}{rcl}\left(n-1\right)\text{E}\left(S_n^2\right)&=&\text{E}\left(\sum _{i=1}^{n}\left(X_1-\bar{X}_n\right)^2\right) \\ &=&\text{E}\left(\sum _{i=1}^{n}X_i^2\right)-\text{E}\left(\sum _{i=1}^{n}2X_i\bar{X}_n\right)+\text{E}\left(\sum _{i=1}^{n}\bar{X}_n^2\right) \\ &=&\text{E}\left(\sum _{i=1}^{n}X_i^2\right)-\text{E}\left(2\bar{X}_n\sum _{i=1}^{n}X_i\right)+\text{E}\left(\bar{X}_n^2\sum _{i=1}^{n}1\right) \\ &=& \text{E}\left(\sum _{i=1}^nX_i^2\right)-\text{E}\left(2n\bar{X}_n\right)^2+\text{E}\left(n\bar{X}_n^2\right) \\ &=& \text{E}\left(\sum _{i=1}^{n} X_i^2\right)-n\text{E}\left(\bar{X}_n^2\right) \\ &=& n\text{E}\left(X_i^2\right)-n\text{E}\left(\bar{X}_n^2\right) \\ &\Leftrightarrow& \\ \frac{n-1}{n}\text{E}\left(S_n\right)^2&=&\text{E}\left(X_i^2\right)-\text{E}\left(\bar{X}_n^2\right)\end{array}$

$\begin{array}{rcl}\text{E}\left(X_i^2\right)&=&\text{E}\left(X_i\right)^2+\text{V}\left(X\right) \\ &=&\mu^2+\sigma^2\end{array}$

$\begin{array}{rcl}\text{E}\left(\bar{X}_n^2\right)&=&\text{E}\left(\bar{X}_n\right)^2+\text{V}\left(\bar{X}_n\right) \\ &=&\mu^2+\text{V}\left(\frac{1}{n}\sum _{i=1}{n}X_i\right) \\ &=&\mu^2 + \frac{1}{n^2}\text{V}\left(\sum _{i=1}^{n}X_i\right) \\ &=& \mu^2+\frac{\sigma^2}{n}\end{array}$