極値分布の基本情報

ガンベル分布


$\begin{align*}f(x) = \frac{1}{\theta} \exp\left(-\frac{x-\mu}{\theta}\right) \exp\left(-\exp\left(-\frac{x-\mu}{\theta}\right)\right)\end{align*}$

期待値:
$E[X] = \mu + \gamma\theta\quad$( $\gamma$ :オイラー・マスケローニ定数)

分散:
$\begin{align*}V[X] = \frac{\pi^2\theta^2}{6}\end{align*}$

母関数:
$\begin{align*}M(t) = \exp(\mu t) \Gamma(1-\theta t)\quad(t < \frac{1}{\theta}) \end{align*}$

フレシェ分布


$\begin{align*}f(x) = \frac{\alpha}{\theta} \left( \frac{x-\mu}{\theta} \right)^{-(1 + \alpha)} \exp\left(-\left( \frac{x-\mu}{\theta} \right)^{-\alpha} \right)\end{align*}$

期待値:
$\begin{align*}E[X] = \mu + \theta \Gamma\left(1-\frac{1}{\alpha}\right)\quad (\alpha > 1)\end{align*}$

分散:
$\begin{align*}V[X] = \theta^2 \left( \Gamma\left(1-\frac{2}{\alpha}\right)-\left(\Gamma\left(1-\frac{1}{\alpha}\right)\right)^2 \right)\quad(\alpha > 2)\end{align*}$

母関数:
$\alpha > k$ ならば $k$ 次モーメントが存在

ワイブル分布


$\begin{align*}f(x) = \frac{\alpha}{\theta} \left( \frac{x}{\theta} \right)^{\alpha-1} \exp\left(-\left( \frac{x}{\theta} \right)^\alpha \right)\end{align*}$

期待値:
$\begin{align*}E[X] = \theta \Gamma\left(1 + \frac{1}{\alpha}\right)\end{align*}$

分散:
$\begin{align*}V[X] = \theta^2 \left( \Gamma\left(1 + \frac{2}{\alpha}\right)-\left(\Gamma\left(1 + \frac{1}{\alpha}\right)\right)^2 \right)\end{align*}$

母関数:
$\begin{align*}M(t) = \sum_{n=0}^{\infty} \frac{(t \theta)^n}{n!} \Gamma\left( 1 + \frac{n}{\alpha} \right)\end{align*}$

タイトルとURLをコピーしました