axjack's blog

### axjack is said to be an abbreviation for An eXistent JApanese Cool Klutz ###

統計学実践ワークブック第3章:分布の特性値〜繰り返し期待値の法則および繰り返し分散の法則〜

統計学実践ワークブックのp.17です。ふわっと証明します。

繰り返し期待値の法則

 \displaystyle

E[ Y  ] = E[ E[ Y | X ] ]

証明

 
\displaystyle  E[ E[ Y|X ]  = E^X [  E^{Y|X} [ Y|X ]  ] \\
\displaystyle  = E^X[ \sum_y y P(Y=y | X= x) ] \\
\displaystyle  = \sum_x \sum_y y P(Y=y | X= x) P(X=x) \\
\displaystyle  = \sum_x \sum_y y P(Y=y , X=x ) \\
\displaystyle  =  \sum_y y  \sum_x P(Y=y , X=x ) \\
\displaystyle  =  \sum_y y  P(Y=y) \\
\displaystyle  =  E[ Y ] \\

繰り返し分散の法則

 
\displaystyle V[ Y ]  = E[ V[ X|Y ] ] + V[ E[ X|Y ]]

証明


\displaystyle  V\left[ Y  \right] = E\left[ Y^2 \right] - \left(E\left[ Y \right]\right)^2 \\
\displaystyle = E\left[ E\left[ Y^2 | X \right] \right] - \left(E\left[ E \left[ Y| X \right] \right]  \right) ^2 \\
\displaystyle = E\left[ V\left[ Y|X \right] + \left(E\left[ Y | X \right]\right)^2 \right] - \left(E\left[ E \left[ Y| X \right] \right]  \right) ^2 \\
\displaystyle = E\left[ V\left[ Y|X \right] \right]  +  E\left[  \left(E\left[ Y | X \right]\right)^2 \right] - \left(E\left[ E \left[ Y| X \right] \right]  \right) ^2 \\
\displaystyle = E\left[ V\left[ Y|X \right] \right]  +  V\left[  E\left[ Y | X \right] \right]

axjack is said to be an abbreviation for An eXistent JApanese Cool Klutz.