\documentclass[全部作业]{subfiles} \begin{document} \chapter{单元作业4} \begin{enumerate} \item 若二维离散型随机变量$(X,Y)$有联合分布列如下: \begin{center} % \renewcommand\arraystretch{1.5} \begin{tabular}{c|ccc} % \hline \diagbox{$X$}{$Y$} & $0$ & $1$ & $2$ \\ \hline $0$ & $\frac{1}{2}$ & $\frac{1}{8}$ & $\frac{1}{4}$ \\ % \hline $1$ & $\frac{1}{16}$ & $\frac{1}{16}$ & $0$ \\ % \hline \end{tabular} \end{center} 求$X$与$Y$的边际分布列。 \begin{proof}[解] \begin{zhongwen} $X$和$Y$的边际分布列分别为 \begin{center} \renewcommand\arraystretch{1.5} \begin{tabular}{c|cc} $X$ & $0$ & $1$ \\ \hline $P$ & \makecell{$\dfrac{7}{8}$} & $\dfrac{1}{8}$ \\ \end{tabular} \qquad \begin{tabular}{c|ccc} $Y$ & $0$ & $1$ & $2$ \\ \hline $P$ & $\dfrac{9}{16}$ & $\dfrac{3}{16}$ & $\dfrac{1}{4}$ \\ \end{tabular} \end{center} \end{zhongwen} \end{proof} \item 设随机变量$(X,Y)$的概率密度函数为 $$ p(x,y)=\begin{cases} 3x,\quad & \text{若}01 \\ \end{cases}\\ &=\begin{cases} 3x^{2},\quad & 0\leqslant x\leqslant 1 \\ 0,\quad & x<0 或x>1 \\ \end{cases} \end{aligned} $$ $$ \begin{aligned} p_{Y}(y)&=\int_{-\infty}^{+\infty} p(x,y) \mathrm{d}x=\begin{cases} \int_{y}^{1} 3x \mathrm{d}x, \quad & 0\leqslant y\leqslant 1 \\ 0,\quad & x<0 或 x>1 \\ \end{cases}\\ &=\begin{cases} \frac{3}{2}-\frac{3}{2}y^{2},\quad & 0\leqslant y\leqslant 1 \\ 0,\quad & y<0 或 y>1 \\ \end{cases} \end{aligned} $$ \end{window} \noindent$p(1,1)=3,p_{X}(1)=3,p_{Y}(1)=0$,于是$p(1,1)\neq p_{X}(1)\cdot p_{Y}(1)$,所以$X$和$Y$互相不独立。 \end{zhongwen} \end{proof} \item 设随机变量$X$与$Y$相互独立,且$X$服从均匀分布$U(0,1)$,$Y$服从指数分布$\operatorname{Exp}(1)$。求 \begin{enumerate} \item $(X,Y)$的联合概率密度函数$p(x,y)$; \begin{proof}[解] \begin{zhongwen} $$ p_{X}(x)=\begin{cases} 1,\quad & x\in [0,1] \\ 0,\quad & x<0 或 x>1 \\ \end{cases}\quad p_{Y}(y)=\begin{cases} e^{-y},\quad & y\geqslant 0 \\ 0,\quad & y< 0 \\ \end{cases} $$ $$ p(x,y)=\begin{cases} e^{-y},\quad & x\in [0,1], y\geqslant 0 \\ 0,\quad & 其他 \\ \end{cases} $$ \end{zhongwen} \end{proof} \item 概率$P(X+Y\leqslant 1)$; \begin{proof}[解] \begin{zhongwen} 令$Z=X+Y$,则随机变量$Z$的概率密度函数为 $$ p_{Z}(z)=\int_{-\infty}^{+\infty} p_{X}(x)p_{Y}(z-x) \mathrm{d}x=\begin{cases} \int_{0}^{1} e^{x-z} \mathrm{d}x ,\quad & z>1\\ \int_{0}^{z} e^{x-z} \mathrm{d}x,\quad & 0\leqslant z\leqslant 1 \\ 0,\quad & z<0 \\ \end{cases}=\begin{cases} e^{1-z}-e^{-z},\quad & z>1 \\ 1-e^{-z},\quad & 0\leqslant z\leqslant 1 \\ 0,\quad & z<0 \\ \end{cases} $$ 所以 $$ P(X+Y\leqslant 1)=\int_{-\infty}^{1} p_{Z}(z) \mathrm{d}z=\int_{0}^{1} (1-e^{-z}) \mathrm{d}z=1+(\frac{1}{e}-1)=\frac{1}{e} $$ \end{zhongwen} \end{proof} \item 概率$P(X\leqslant Y)$。 \begin{proof}[解] \begin{zhongwen} 令随机变量$Z=X-Y$,则$Z$的概率密度函数为 $$ p_{Z}(z)=\int_{-\infty}^{+\infty} p(x,x-z) \mathrm{d}x=\begin{cases} 0,\quad & z>1 \\ \int_{z}^{1} e^{z-x} \mathrm{d}x,\quad & 0\leqslant z\leqslant 1 \\ \int_{0}^{1} e^{z-x} \mathrm{d}x,\quad & z<0 \\ \end{cases}=\begin{cases} 0,\quad & z>1 \\ 1-e^{z-1},\quad & 0\leqslant z\leqslant 1 \\ e^{z}-e^{z-1},\quad & z<0 \\ \end{cases} $$ 所以 $$ P(X\leqslant Y)=P(X-Y\leqslant 0)=\int_{-\infty}^{0} (e^{z}-e^{z-1}) \mathrm{d}z=1-\frac{1}{e} $$ \end{zhongwen} \end{proof} \end{enumerate} \item 设随机变量$(X,Y)$的概率密度函数为 $$ p(x,y)=\begin{cases} Ae^{-(3x+2y)},\quad & x>0,y>0; \\ 0,\quad & \text{otherwise}. \\ \end{cases} $$ 求: \begin{enumerate} \item 常数$A$的值; \begin{proof}[解] \begin{zhongwen} $$ \begin{aligned} &\int_{-\infty}^{+\infty} \int_{-\infty}^{+\infty} p(x,y) \mathrm{d}x \mathrm{d}y=\int_{0}^{+\infty} \int_{0}^{+\infty} Ae^{-(3x+2y)} \mathrm{d}x \mathrm{d}y=\int_{0}^{+\infty} \left. -\frac{1}{3}Ae^{-(3x+2y)} \right|_{0}^{+\infty} \mathrm{d}y\\ =&\int_{0}^{+\infty} \frac{1}{3}Ae^{-2y} \mathrm{d}y=\left. -\frac{1}{6}Ae^{-2y} \right|_{0}^{+\infty}=\frac{1}{6}A=1 \end{aligned} $$ 所以常数$A$的值为$6$。 \end{zhongwen} \end{proof} \item 分布函数$F(x,y)$; \begin{proof}[解] \begin{zhongwen} 由题意可知,当$x\leqslant 0$ 或$y\leqslant 0$时,$F(x,y)=0$。 当$x>0,y>0$时,$ \displaystyle F(x,y) =\int_{0}^{y} \int_{0}^{x} 6e^{-(3u+2v)} \mathrm{d}u \mathrm{d}v =\int_{0}^{y} \left. -2e^{-(3u+2v)} \right|_{0}^{x} \mathrm{d}v =\int_{0}^{y} \left(2e^{-2v}-2e^{-(3x+2v)}\right) \mathrm{d}v =\left. -e^{-2v} \right|_{0}^{y}+\left. e^{-(3x+2v)} \right|_{0}^{y} =-e^{-2y}+1+e^{-(3x+2y)}-e^{-3x} =e^{-(3x+2y)}-e^{-3x}-e^{-2y}+1 $ 所以 $$ \begin{aligned} F(x,y)=\begin{cases} e^{-(3x+2y)}-e^{-3x}-e^{-2y}+1,\quad & x>0,y>0 \\ 0,\quad & \text{otherwise} \\ \end{cases} \end{aligned} $$ \end{zhongwen} \end{proof} \item 概率$P(-2