概率论习题之标准正态绝对值的期望

概率论习题之标准正态绝对值的期望

一、主要注意的点

$$\begin{gathered} E|X|=\sqrt{\frac{2}{\Pi }} \\ 计算:E|X|={\int }_{-\infty }^{+\infty }|X|f(x)dx \\ EZ=E|x-\mu | \end{gathered}$$

二、习题

$X\sim N(0,1),E(|X|)=\sqrt{\frac{2}{\Pi }}$

$X\sim N(3,9),E(|X-3|)=$
处理一般正态的基本操作：1.标准化 2.标准正态
$$\begin{gathered} E|\frac{X-3}{3}|=\sqrt{\frac{2}{\pi }} \\ 等式左右两边同时乘以3 \\ E|X-3|=\frac{3\sqrt{2}}{\sqrt{\pi }} \end{gathered}$$

$X\sim N(\mu ,{\sigma }^2),Z=|X-\mu |,E(Z)=$
$$\begin{gathered} 标准化:E|\frac{X-\mu }{\sigma }|=\frac{2}{\pi } \\ 默认\sigma 大于0，即在等式两边同时乘以\sigma \\ E|X-\mu |=\frac{\sqrt{2}\mu }{\sqrt{\pi }} \end{gathered}$$

设（ x , y ) ∼ N ( 0 , 0 ; 1 2 , 1 2 ; 0 ) , ϕ ( x ) 为标准正态函数，则 P { X − Y < E ∣ X − Y ∣ } = S E T   ( x , y ) ∼ N ( 0 , 0 ; 1 2 , 1 2 ; 0 ) , ϕ ( x )   i s   a   s t a n d a r d   n o r m a l   f u n c t i o n ,   t h e n   P { X − Y < E ∣ X − Y ∣ } = 设（x,y)\sim N(0,0;\frac{1}{2},\frac{1}{2};0),\phi(x)为标准正态函数，则P \lbrace X-Y<E|X-Y|\rbrace=\\ SET ~(x,y)\sim N(0,0; \frac{1}{2},\frac{1}{2}; 0), \phi(x) ~is ~a ~standard~ normal ~function, ~then ~P \lbrace X-Y<E|X-Y|\rbrace= 设（x,y)∼N(0,0;21​,21​;0),ϕ(x)为标准正态函数，则P{X−Y<E∣X−Y∣}=SET (x,y)∼N(0,0;21​,21​;0),ϕ(x) is a standard normal function, then P{X−Y<E∣X−Y∣}=
$$\begin{gathered} X\sim N(0,\frac{1}{2}) \\ Y\sim N(0,\frac{1}{2}) \\ (正态分布的线性组合服从一维正态) \\ Linearcombinations \\ ofnormaldistributionsobeyone-dimensionalnormal \\ X-Y\sim N(0,1)(期望相减，方差相加) \\ Expectationsaresubtracted,andvariancesareadded \\ E(X-Y)=\frac{\sqrt{2}}{\sqrt{\pi }} \\ pleaseletZ=x-y \\ SoPZ<E|X-Y|=\Phi (\sqrt{\frac{2}{\pi }}) \end{gathered}$$

$$\begin{gathered} 设X1，X2为来自总体N(\mu ,{\sigma }^2)的简单随机样本，记\hat{\sigma }=a|X1-X2|,若E(\sigma )=\sigma ,则a= \\ LetX1andX2besimplerandomsamplesfromthepopulationN(\mu ,{\sigma }^2)anddenote\hat{\sigma }=a|X1-X2|, \\ ifE(\sigma )=\sigma ,thena= \end{gathered}$$
$$\begin{gathered} X1\sim N(\mu ,{\sigma }^2) \\ X2\sim N(\mu ,{\sigma }^2) \\ E(\hat{\sigma })=E(a|X1-X2|)=aE|X1-X2| \\ X1-X2\sim N(0,2{\sigma }^2) \\ \frac{X1-X2}{\sqrt{2}\mu }\sim N(0,1) \\ E|\frac{X1-X2}{\sqrt{2}\mu }=\sqrt{\frac{2}{\pi }} \\ Bothsidesoftheequationaremultipliedby\sqrt{2}\mu atthesametime \\ E|X1-X2|=\frac{2\sigma }{\sqrt{\pi }} \\ SoE(\hat{\sigma })=a\cdot \frac{2\sigma }{\sqrt{\pi }}=\sigma \\ \frac{2}{\sqrt{\pi }}\cdot a=1 \\ a=\frac{\sqrt{\pi }}{2} \end{gathered}$$

设二维随机变量 ( X 1 , X 2 ) ∼ N ( 0 , 0 ; 1 , 1 ; 0 ) , 记 X = max ⁡ { X 1 , X 2 } , Y = min ⁡ { X 1 , X 2 } , Z = X − Y , 求 E ( Z ) L e t   t h e   t w o − d i m e n s i o n a l   r a n d o m   v a r i a b l e ( X 1 , X 2 ) ∼ N ( 0 , 0 ; 1 , 1 ; 0 ) ,   w r i t e   X = m a x { X 1 , X 2 } , Y = m i n { X 1 , X 2 } , Z = X − Y , a n d   f i n d E ( Z ) 设二维随机变量(X1,X2)\sim N(0,0;1,1;0),记X=\max\lbrace X1,X2 \rbrace,Y=\min\lbrace X1,X2 \rbrace,Z=X-Y,求E(Z)\\ Let ~the~ two-dimensional ~random ~variable(X1,X2)\sim N(0,0; 1,1; 0),\\ ~ write ~X=max\lbrace X1,X2 \rbrace,Y=min\lbrace X1,X2 \rbrace,Z=X-Y, and ~find E(Z) 设二维随机变量(X1,X2)∼N(0,0;1,1;0),记X=max{X1,X2},Y=min{X1,X2},Z=X−Y,求E(Z)Let the two−dimensional random variable(X1,X2)∼N(0,0;1,1;0), write X=max{X1,X2},Y=min{X1,X2},Z=X−Y,and findE(Z)
$$\begin{gathered} X1\sim N(0,1) \\ X2\sim N(0,1) \\ E(Z)=E(X-Y) \\ ifX1>X2,Z=X-Y=X1-X2 \\ elseZ=X-Y=X2-X1 \\ SoZ=|X1-X2| \\ E(Z)=E(|X-Y|) \\ standardization:\frac{X1-X2}{\sqrt{2}}\sim N(0,1) \\ E|\frac{X1-X2}{\sqrt{2}}|=\sqrt{\frac{2}{\pi }} \\ Bothsidesoftheequationaremultipliedby\sqrt{2}atthesametime \\ E(X1-X2)=\frac{2}{\sqrt{\pi }} \end{gathered}$$

$$\begin{gathered} 设X1与X2相互独立，且服从正态分布N(\mu ,{\sigma }^2),求E\{(X1,X2\}) \\ LetX1andX2beindependentofeachotherandobeythenormaldistributionN(\mu ,{\sigma }^2),andfindE\{(X1,X2\}) \end{gathered}$$
X 1 ∼ N ( μ , σ 2 ) X 2 ∼ N ( μ , σ 2 ) E { max ⁡ ( x 1 , x 2 ) = 1 2 ( x 1 + x 2 + ∣ x 1 − x 2 ∣ ) ) } = 1 2 ( E ( X 1 ) + E ( X 2 ) + E ∣ X 1 − X 2 ∣ ) X 1 − X 2 ∼ N ( 0 , 2 σ 2 ) X 1 − X 2 2 σ ∼ N ( 0 , 1 ) E ∣ X 1 − X 2 2 σ ∣ = 2 π E ( X 1 − X 2 ) = 2 σ π E { max ⁡ ( x 1 , x 2 ) = 1 2 ( x 1 + x 2 + ∣ x 1 − x 2 ∣ ) ) } = 1 2 ( E ( X 1 ) + E ( X 2 ) + E ∣ X 1 − X 2 ∣ ) = 1 2 ( μ + μ + 2 σ π ) = μ + σ π X1\sim N(\mu,\sigma^2)\\ X2\sim N(\mu,\sigma^2)\\ E\lbrace\max(x1,x2)=\frac{1}{2}(x1+x2+|x1-x2|))\rbrace\\ =\frac{1}{2}(E(X1)+E(X2)+E|X1-X2|)\\ X1-X2\sim N(0,2\sigma^2)\\ \frac{X1-X2}{\sqrt{2}\sigma}\sim N(0,1)\\ E|\frac{X1-X2}{\sqrt{2}\sigma}|=\sqrt{\frac{2}{\pi}}\\ E(X1-X2)=\frac{2\sigma}{\sqrt{\pi}}\\ E\lbrace\max(x1,x2)=\frac{1}{2}(x1+x2+|x1-x2|))\rbrace\\ =\frac{1}{2}(E(X1)+E(X2)+E|X1-X2|)\\ =\frac{1}{2}(\mu+\mu+\frac{2\sigma}{\sqrt{\pi}})\\ =\mu+\frac{\sigma}{\sqrt{\pi}}\\ X1∼N(μ,σ2)X2∼N(μ,σ2)E{max(x1,x2)=21​(x1+x2+∣x1−x2∣))}=21​(E(X1)+E(X2)+E∣X1−X2∣)X1−X2∼N(0,2σ2)2 ​σX1−X2​∼N(0,1)E∣2 ​σX1−X2​∣=π2​ ​E(X1−X2)=π ​2σ​E{max(x1,x2)=21​(x1+x2+∣x1−x2∣))}=21​(E(X1)+E(X2)+E∣X1−X2∣)=21​(μ+μ+π ​2σ​)=μ+π ​σ​