Section03_连续型随机变量及其分布

基本概念

  1. 定义FX(x)=P{Xx}=xf(t)dt F_{X}(x) = \mathbb{P}\{X\le x\} = \int_{-\infty}^{x} f(t)\cdot dt
    • 则称XX为连续型变量,f(x)f(x)XX的概率密度函数
    • 注1
      1. 连续型随机变量的分布函数FX(x)F_{X}(x)必定连续
      2. XX的分布函数F(x)F(x)连续且至多有有限个不可导点,则XX为连续型随机变量
    • 注2 F(x)=f(x)F'(x) = f(x) (其中xxf(x)f(x)的连续点)
  2. f(x)f(x)为密度 \Leftrightarrow {f(x)0f(x)dx=1\begin{cases} f(x)\ge 0 \\ \int_{-\infty}^{\infty}f(x)\cdot dx =1 \end{cases}
    • f1(x)={12,0<x<20,其他f_{1}(x) = \begin{cases} \frac{1}{2},& 0<x<2 \\ 0, & \text{其他} \end{cases}f2(x)={12,0x20,其他f_{2}(x) = \begin{cases} \frac{1}{2},& 0\le x\le 2 \\ 0, & \text{其他} \end{cases} 两者均为密度,且f1(x)=f2(x)\color{#D0104C}f_{1}(x) = f_{2}(x)在概率论中对二者不加以区分
  3. 如何由F(x)F(x)f(x)f(x)? 尽管求导,如 F(x)={0,x<0sinx,0x<π21,xπ2X’s Density: f(x)={cosx,0<x<π20,others \begin{array}{ll} & F(x) = \begin{cases} 0, & x<0 \\ \sin x, & 0\le x< \frac{\pi}{2} \\ 1, & x\ge \frac{\pi}{2} \end{cases} \\ \Rightarrow & X \text{'s Density: } f(x) = \begin{cases} \cos x, & 0 < x < \frac{\pi}{2} \\ 0, & \text{others} \end{cases} \\ \end{array}
  4. 求概率
    1. P{X=x0}=F(x0)F(x00)=0\mathbb{P}\{X=x_{0}\} = F(x_{0}) - F(x_{0}-0) = 0
    2. P{a<xb}=P{a<X<b}=P{aXb}=F(b)F(a)=abf(x)dx\displaystyle \mathbb{P}\{a<x\le b\} = \mathbb{P}\{a <X<b\} = \mathbb{P}\{a\le X\le b\} = F(b)-F(a) = \int_{a}^{b}f(x)\cdot dx
    3. 连续型随机变量计算时仅考虑f(x)>0f(x)>0的部分

例题

  1. 设随机变量 XX 的概率密度为 f(x)={16,2<x<013,1<x<30,其他f(x) = \begin{cases} \frac{1}{6}, & -2 < x < 0 \\ \frac{1}{3}, & 1 < x < 3 \\ 0, & \text{其他} \end{cases}P{X25}\mathbb{P}\{X^{2}\le 5\} f(x)={16,2<x<013,1<x<30,其他P{x25}=20f(x)dx+15f(x)dx=16×2+13×(51)=53 \begin{array}{ll} \because & f(x) = \begin{cases} \frac{1}{6}, & -2 < x < 0 \\ \frac{1}{3}, & 1 < x < 3 \\ 0, & \text{其他} \end{cases} \\ \therefore & \displaystyle \mathbb{P}\{x^{2}\le 5\} = \int_{-2}^{0}f(x)\cdot dx + \int_{1}^{\sqrt{5}}f(x)\cdot dx \\ & = \frac{1}{6}\times 2 + \frac{1}{3}\times (\sqrt{5}-1) = \frac{\sqrt{5}}{3} \end{array}
  2. 设随机变量 XX 的概率密度函数为 f(x)={kcosx,x<π20,xπ2f(x) = \begin{cases} k\cos x, & \vert x \vert < \frac{\pi}{2} \\ 0, & \vert x \vert \ge \frac{\pi}{2} \end{cases}
    1. 求常数 kk
    2. XX 的分布函数 F(x)F(x) f(x)dx=1f(x)={kcosx,x<π20,xπ2f(x)dx=π2π2f(x)dx=ksinxπ2π2=2kk=12F(x)=P{Xx}=xf(x)dxF(x)={0,x<π212sinx+12,π2x<π21,xπ2 \begin{array}{ll} \because & \displaystyle \int_{-\infty}^{\infty}f(x)\cdot dx = 1 \\ \because & f(x) = \begin{cases} k\cos x, & \vert x \vert < \frac{\pi}{2} \\ 0, & \vert x \vert \ge \frac{\pi}{2} \end{cases}\\ \therefore & \displaystyle \int_{-\infty}^{\infty}f(x)\cdot dx = \int_{-\frac{\pi}{2}}^{\frac{\pi}{2}}f(x)\cdot dx = k\sin x\vert^{\frac{\pi}{2}}_{-\frac{\pi}{2}} = 2k \\ \therefore & k = \frac{1}{2} \\ \\ & \displaystyle F(x) = \mathbb{P}\{X\le x\} = \int_{-\infty}^{x}f(x)\cdot dx \\ \therefore & F(x) = \begin{cases} 0, & x<-\frac{\pi}{2} \\ \frac{1}{2}\sin x + \frac{1}{2}, & -\frac{\pi}{2}\le x < \frac{\pi}{2} \\ 1, & x\ge \frac{\pi}{2} \end{cases} \\ \end{array}

常见连续型随机变量

  1. 均匀分布 XU(a,b)X\sim U(a,b) Xf(x)={1ba,a<x<b0,其他 X\sim f(x) = \begin{cases} \frac{1}{b-a}, & a < x < b \\ 0, & \text{其他} \end{cases}
    • (c,d)(a,b)(c,d)\subset (a,b),则 P{c<X<d}=dcba\mathbb{P}\{c<X<d\} = \frac{d-c}{b-a}
    • A,BA,B为任意两个事件,分别判断下列命题的正确性
      1. P(AB)=0\mathbb{P}(AB)= 0,则 A,BA,B 互斥 ×XU(1,1),A={1<X0},B={0X<1}X\sim U(-1,1), A =\{-1<X\le 0\}, B =\{0 \le X < 1\}
      2. P(AB)=A\mathbb{P}(AB) = A,则 ABA\subset B ×XU(1,1),A={12<X0},B={12X<0}X\sim U(-1,1), A =\{-\frac{1}{2}<X\le 0\}, B =\{\frac{1}{2} \le X < 0\}
    • 由概率无法推出事件之间的关系(包含,相容,互斥)
  2. 指数分布 XE(λ),λ>0X\sim E(\lambda), \lambda > 0
    1. Xf(x)={λeλx,x>00,x0X\sim f(x) = \begin{cases} \lambda e^{-\lambda x}, & x>0 \\ 0, & x\le 0 \end{cases}
    2. XF(x)=P{Xx}=xf(t)dt={0,x00xλeλtdt=1eλx\displaystyle X\sim F(x) = \mathbb{P}\{X\le x\} = \int_{-\infty}^{x} f(t)\cdot dt = \begin{cases} 0, x \le 0 \\ \displaystyle \int_{0}^{x}\lambda e^{-\lambda t}\cdot dt \\ \end{cases} = 1-e^{-\lambda x}
    3. 相关问题 寿命 P{X>a}={1P{Xa}=1F(a)=eλaa+λeλxdx=eλxa+=eλa\mathbb{P}\{X > a\} = \begin{cases} 1- \mathbb{P}\{X\le a\} = 1-F(a) = e^{-\lambda a} \\ \displaystyle \int_{a}^{+\infty} \lambda e^{-\lambda x}\cdot dx = -e^{-\lambda x}\vert_{a}^{+\infty} = e^{-\lambda a} \end{cases}
    4. 无记忆性s>0,t>0s>0, t>0P{X>s+tX>s}=P{X>s+t,X>s}P{X>s}=P{X>s+t}P{X>s}=eλ(s+t)eλs=eλt=P{X>t} \begin{array}{ll} & \displaystyle \mathbb{P}\{X>s+t\vert X>s\} = \frac{\mathbb{P}\{X > s+t, X > s\}}{\mathbb{P}\{X > s\}} \\ & \displaystyle = \frac{\mathbb{P}\{X > s + t\}}{\mathbb{P}\{X>s\}} = \frac{e^{-\lambda(s+t)}}{e^{-\lambda s}} = e^{-\lambda t} = \mathbb{P}\{X> t\} \end{array}
  3. 正态分布 XN(μ,σ2),σ>0X\sim \mathcal{N}(\mu,\sigma^{2}), \sigma>0
    1. Xf(x)=12πσe(xμ)22σ2,xRX\sim f(x) = \frac{1}{\sqrt{2\pi}\sigma}e^{-\frac{(x-\mu)^{2}}{2\sigma^{2}}}, \quad x\in \mathbb{R}
    2. F(x)=x12πσe(tμ)22σ2dt\displaystyle F(x) = \int_{-\infty}^{x}\frac{1}{\sqrt{2\pi}\sigma}e^{-\frac{(t - \mu)^{2}}{2\sigma^{2}}} \cdot dt 无法积分;但由 F(x)=f(x)>0F'(x) = f(x) >0 可得,F(x)F(x) 单调递增
    3. Y=XμσN(0,1)Y = \frac{X-\mu}{\sigma}\sim \mathcal{N}(0,1)
      • YY 的密度 φ(y)=12πey22,yR\varphi(y) = \frac{1}{\sqrt{2\pi}}e^{-\frac{y^{2}}{2}}, \quad y\in \mathbb{R}
      • YY 的分布函数 Φ(y)=x12πet22dt\displaystyle \Phi(y) = \int_{-\infty}^{x}\frac{1}{\sqrt{2\pi}}e^{-\frac{t^{2}}{2}}\cdot dt
        • Φ(0)=0.5\Phi(0) = 0.5
        • Φ(1)=0.8413\Phi(1) = 0.8413
        • Φ(1.645)=0.95\Phi(1.645) = 0.95
        • Φ(1.96)=0.975\Phi(1.96) = 0.975
    4. P{a<X<b}=P{aμσ<Y<bμσ}=Φ(bμσ)Φ(aμσ)\mathbb{P}\{a<X<b\} = \mathbb{P}\{\frac{a-\mu}{\sigma} < Y < \frac{b-\mu}{\sigma}\} = \Phi(\frac{b-\mu}{\sigma}) - \Phi(\frac{a-\mu}{\sigma})
    5. +12πσe(xμ)22σ2dx=1\displaystyle \int_{-\infty}^{+\infty}\frac{1}{\sqrt{2\pi}\sigma}e^{-\frac{(x-\mu)^{2}}{2\sigma^{2}}}\cdot dx = 1 的妙用,凑!
      • 0+ex2dx=12+ex2dx=12+ex22×12π2+12π×12ex22×12=π2 \begin{array}{ll} & \displaystyle \int_{0}^{+\infty}e^{-x^{2}}\cdot dx = \frac{1}{2}\int_{-\infty}^{+\infty} e^{-x^{2}}\cdot dx = \frac{1}{2} \int_{-\infty}^{+\infty} e^{-\frac{x^{2}}{2\times \frac{1}{2}}} \\ \therefore & \displaystyle \frac{\sqrt{\pi}}{2}\int_{-\infty}^{+\infty}\frac{1}{\sqrt{2\pi}\times \sqrt{\frac{1}{2}}} e^{-\frac{x^{2}}{2\times \frac{1}{2}}} = \frac{\sqrt{\pi}}{2} \\ \end{array}

results matching ""

    No results matching ""