标准正态分布k阶原点矩公式
今天在看
χ
2
\chi^2
χ2分布,计算其方差时,遇到了标准正态分布的四阶原点矩。书上直接写
E
(
X
i
4
)
=
3
E(X_i^4)=3
E(Xi4)=3,很好奇。设
X
i
∼
N
(
0
,
1
)
X_i\sim\mathcal{N}(0,1)
Xi∼N(0,1)想根据定义计算:
E
(
X
i
4
)
=
∫
−
∞
+
∞
x
4
1
2
π
e
−
x
2
2
d
x
E(X_i^4)=\int_{-\infty}^{+\infty}x^4\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}}dx
E(Xi4)=∫−∞+∞x42π1e−2x2dx
计算起来复杂度有点高,K阶就更不敢想。
想着估计结果是
k
k
k的递推关系式。所有直接计算:
E
(
X
k
)
=
∫
−
∞
+
∞
x
k
1
2
π
e
−
x
2
2
d
x
=
1
k
+
1
x
k
+
1
1
2
π
e
−
x
2
2
∣
−
∞
+
∞
+
1
k
+
1
∫
−
∞
+
∞
x
k
+
2
1
2
π
e
−
x
2
2
d
x
=
0
+
1
k
+
1
E
(
X
k
+
2
)
\begin{aligned} E(X^k) &= \int_{-\infty}^{+\infty}x^k\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}}dx \\ & = \frac{1}{k+1}x^{k+1}\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}} |_{-\infty}^{+\infty} + \frac{1}{k+1}\int_{-\infty}^{+\infty}x^{k+2}\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}}dx \\ & = 0 + \frac{1}{k+1}E(X^{k+2}) \end{aligned}
E(Xk)=∫−∞+∞xk2π1e−2x2dx=k+11xk+12π1e−2x2∣−∞+∞+k+11∫−∞+∞xk+22π1e−2x2dx=0+k+11E(Xk+2)
从上式看,结果已经很明显了,其递推关系为:
E
(
X
k
)
=
(
k
−
1
)
E
(
X
k
−
2
)
,
k
=
2
,
3
,
4
⋯
E(X^k) = (k-1)E(X^{k-2}),k=2,3,4\cdots
E(Xk)=(k−1)E(Xk−2),k=2,3,4⋯
其中:
E
(
X
0
)
=
∫
−
∞
+
∞
x
0
1
2
π
e
−
x
2
2
d
x
=
1
E(X^0)=\int_{-\infty}^{+\infty}x^0\frac{1}{\sqrt{2\pi}}e^{-\frac{x^2}{2}}dx=1
E(X0)=∫−∞+∞x02π1e−2x2dx=1
E
(
X
1
)
=
0
E(X^1)=0
E(X1)=0
所以,当
k
=
2
i
k=2i
k=2i为偶数时:
E
(
X
k
)
=
(
k
−
1
)
(
k
−
3
)
⋯
3
×
1
×
E
(
X
0
)
=
∏
i
=
1
k
/
2
(
2
i
−
1
)
\begin{aligned} E(X^k) &= (k-1)(k-3)\cdots3\times1\times E(X^0) \\ &=\prod_{i=1}^{k/2}(2i-1) \end{aligned}
E(Xk)=(k−1)(k−3)⋯3×1×E(X0)=i=1∏k/2(2i−1)
当
k
=
2
i
−
1
k=2i-1
k=2i−1为奇数时:
E
(
X
k
)
=
(
k
−
1
)
(
k
−
3
)
⋯
3
×
1
×
E
(
X
1
)
=
0
\begin{aligned} E(X^k) &= (k-1)(k-3)\cdots3\times1\times E(X^1) \\ &=0 \end{aligned}
E(Xk)=(k−1)(k−3)⋯3×1×E(X1)=0
综上有:
E
(
X
k
)
=
{
∏
i
=
1
k
/
2
(
2
i
−
1
)
k
=
2
i
,
i
=
1
,
2
,
3
⋯
0
k
=
2
i
−
1
E(X^k) = \left \{ \begin{array}{ll} \prod_{i=1}^{k/2}(2i-1) & k=2i, i=1,2,3\cdots \\ & \\ 0 & k=2i-1 \end{array} \right.
E(Xk)=⎩⎨⎧∏i=1k/2(2i−1)0k=2i,i=1,2,3⋯k=2i−1