1
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 3 years have passed since last update.

2.10(標準) 積率母関数,変数変換 2(1.方針で積率母関数を解説)

Last updated at Posted at 2021-07-17

1.方針

確率変数Xの積率母関数とは,

M_X(t)=\mathbb{E}_{X\sim f_X}[e^{tX}]

というtについての関数である.これを微分し,t=0を代入したものはXの期待値である.

\begin{align}
\left.\frac{d}{dt}M_X(t)\right|_{t=0}&= \left.\frac{d}{dt}\mathbb{E}_{X\sim f_X}[e^{tX}]\right|_{t=0}\\
&= \left.\frac{d}{dt}\int e^{tx}f_X(x)dx\right|_{t=0}\\
&= \left.\int e^{tx} xf_X(x)dx\right|_{t=0}\\
&= \int xf_X(x)dx\\
&= \mathbb{E}_{X\sim f_X}[X]
\end{align}

同様に,これをk階微分し,t=0を代入したものはXのk階モーメントである.

\begin{align}
\left.\frac{d^k}{dt^k}M_X(t)\right|_{t=0}&= \left.\frac{d^k}{dt^k}\mathbb{E}_{X\sim f_X}[e^{tX}]\right|_{t=0}\\
&= \left.\frac{d^k}{dt^k}\int e^{tx}f_X(x)dx\right|_{t=0}\\
&= \left.\int e^{tx} x^kf_X(x)dx\right|_{t=0}\\
&= \int x^kf_X(x)dx\\
&= \mathbb{E}_{X\sim f_X}[X^k]
\end{align}

1.答案

\begin{align}
M_X(t)&= \int_0^\infty e^{tx}e^{-x}dx\\
&= \int_0^\infty e^{x(t-1)}dx\\
&= \left[\frac{1}{t-1}e^{x(t-1)}\right]_0^\infty\\
&= \frac{1}{1-t}
\end{align}

2.方針

1で求めた積率母関数をk階微分することでk次モーメントを求める.答案では1~4階微分まで求めることで法則性を見つけている.

2.答案

\begin{align}
\frac{d}{dt}M_X(t)&= \frac{1}{(1-t)^2}\\
\frac{d^2}{dt^2}M_X(t)&= \frac{2(1-t)}{(1-t)^4}=\frac{2}{(1-t)^3}\\
\frac{d^3}{dt^3}M_X(t)&= \frac{3(1-t)^2\times 2}{(1-t)^6}=\frac{3!}{(1-t)^4}\\
\frac{d^4}{dt^4}M_X(t)&= \frac{4(1-t)^3\times 3!}{(1-t)^8}=\frac{4!}{(1-t)^5}\\
&\vdots\\
\frac{d^k}{dt^k}M_X(t)&= \frac{k(1-t)^{k+1}(k-1)!}{(1-t)^{2(k+1)}}=\frac{k!}{(1-t)^{k+1}}\\
\therefore\mathbb{E}_{X\sim f_X}[X^k]&= \left.\frac{d^k}{dt^k}M_X(t)\right|_{t=0}=k!
\end{align}

3.方針

変数変換の際はまず分布関数から考える.分布関数ができたらそれを微分することで確率密度関数を求める.

3.答案

Yの分布関数は,

\begin{align}
F_Y(y)&= P[Y\leq y]\\
&= P[\sigma X+\mu\leq y]\\
&= P[X\leq \frac{y-\mu}{\sigma}]\ \ (\because\sigma>0)\\
&= \int_0^{\frac{y-\mu}{\sigma}}e^{-x}dx\\
&= \left[-e^{-x} \right]_0^{\frac{y-\mu}{\sigma}}\\
&= -e^{-\frac{y-\mu}{\sigma}}+1\ \ (\mu<y)
\end{align}

なので,Yの確率密度関数は

f_Y(y)=\frac{d}{dy}F_Y(y)=\frac{1}{\sigma}\exp(-\frac{y-\mu}{\sigma})\ \  \ (\mu<y)

参考文献

1
0
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
1
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?