跳转至

正态分布

连续型随机变量 \(X\)概率密度

\[ f\left(x \right)=\frac{1}{\sqrt{2\pi}\sigma} e^{-\tfrac{\left(x-\mu \right)^2}{2\sigma^2}} \ , \ -\infty<x<+\infty \]

其中 \(\mu\)\(\sigma \ \left(\sigma > 0 \right)\) 为常数,则称 \(X\) 服从参数为 \(\mu\)\(\sigma^2\) 的正态分布 (normal distribution) 或高斯分布 (Gaussian distribution),记为 \(X \sim N(\mu, \sigma^2)\)

  • \(f\left(x \right)\) 关于 \(x=\mu\) 对称,在 \(x=\mu\) 处取得最大值 \(f\left(\mu \right)=\dfrac{1}{\sqrt{2\pi}\sigma}\)
  • \(\mu\) 为位置参数。改变 \(\mu\),函数图像将沿 \(x\) 轴平移。
  • \(\sigma\) 越大,图形越扁。\(\sigma\) 越小,图形越尖,\(X\) 落在 \(\mu\) 附近的概率越大。

分布函数为

\[ F\left(x \right) = \int_{-\infty}^{x} \frac{1}{\sqrt{2\pi}\sigma} e^{-\tfrac{\left(t-\mu \right)^2}{2\sigma^2}} \mathrm{d}t \ , \ -\infty<x<+\infty \]
  • \(F\left(\mu \right) = \dfrac{1}{2}\)
  • \(P\left(X \le \mu \right)=P\left(X > \mu \right)=\dfrac{1}{2}\)

标准正态分布

\(X \sim N(\mu, \sigma^2)\),若 \(\mu=0\)\(\sigma^2=1\),则称 \(X\) 服从标准正态分布 (standard normal distribution),记为 \(X \sim N(0, 1)\)

概率密度为

\[ \varphi\left(x \right)=\frac{1}{\sqrt{2\pi}} e^{-\tfrac{x^2}{2}} \ , \ -\infty<x<+\infty \]

分布函数为

\[ \Phi\left(x \right) = \int_{-\infty}^{x} \frac{1}{\sqrt{2\pi}} e^{-\tfrac{t^2}{2}} \mathrm{d}t \ , \ -\infty<x<+\infty \]
  • \(\Phi\left(0 \right) = \dfrac{1}{2}\)
  • \(P\left(X \le 0 \right)=P\left(X > 0 \right)=\dfrac{1}{2}\)
  • \(\Phi\left(-x \right) = 1 - \Phi\left(x \right)\)

\(X \sim N(\mu, \sigma^2)\)

  • \(Z=\dfrac{X-\mu}{\sigma} \sim N(0,1)\)\(Z\)\(X\) 的标准化。
  • \(Y=aX+b \sim N(a\mu+b, (a\sigma)^2)\)\(\left(a \ne 0\right)\)。线性变换后正态性不变。
  • \(F\left(x \right) = \Phi\left(\dfrac{x-\mu}{\sigma} \right)\)
\[ P\left(x_1 < X \le x_2 \right) = \Phi\left(\dfrac{x_2-\mu}{\sigma} \right) - \Phi\left(\dfrac{x_1-\mu}{\sigma} \right) \]

3 sigma 规则

正态分布的随机变量的取值在 \(\mu\)\(3\sigma\) 邻域内的概率为 \(0.9972\),所以该事件的发生几乎是肯定的。

  • \(x > 4\) 时,\(\Phi\left(x \right) \approx 1\)
  • \(x < -4\) 时,\(\Phi\left(x \right) \approx 0\)

相关文章