1. 设X服从期望为2方差为1 的正态分布,则求E(X|X>0)
2. 若X服从参数为10 的t分布,则
X
4
X^4
X4的数学期望为多少?
解:
由题知:X服从n=10的t分布,则X~
Z
Y
/
n
\frac{Z}{\sqrt{Y/n}}
Y/nZ,其中Z服从正态分布Z~N(0,1),Y服从卡方分布Y~
x
2
(
n
)
x^2(n)
x2(n),且Z也服从n=1的卡方分布,Z~
x
2
(
1
)
x^2(1)
x2(1)
则当n=10时, X 2 = Z 2 Y / 10 = Z 2 / 1 Y / 10 X^2 = \frac{Z^2}{Y/10} =\frac{Z^2/1}{Y/10} X2=Y/10Z2=Y/10Z2/1可以发现 X 2 X^2 X2是服从F分布的。表示为 X 2 X^2 X2~ F ( 1 , 10 ) F(1,10) F(1,10)
F分布,当X~ Y / m Z / n \frac{Y/m}{Z/n} Z/nY/m,表示为X~F(m,n)且F分布的均值为当n>2时是 E ( X ) = n n − 2 E(X) = \frac{n}{n-2} E(X)=n−2n,当n>4方差是 D ( X ) = 2 n 2 ( m + n − 2 ) m ( n − 2 ) 2 ( n − 4 ) D(X) =\frac{2n^2(m+n-2)}{m(n-2)^2(n-4)} D(X)=m(n−2)2(n−4)2n2(m+n−2)
所以根据公式
E
X
4
=
E
2
(
X
2
)
+
D
X
2
EX^4=E^2(X^2)+DX^2
EX4=E2(X2)+DX2只需要求出
X
2
X^2
X2的均值和方差即可求得
E
X
4
EX^4
EX4
E
(
X
2
)
=
10
/
(
10
−
2
)
=
5
/
4
E(X^2)=10/(10-2)=5/4
E(X2)=10/(10−2)=5/4
D
(
X
2
)
=
2
∗
1
0
2
(
1
+
10
−
2
)
1
∗
(
10
−
2
)
2
(
10
−
4
)
≈
4.6875
D(X^2) =\frac{2*10^2(1+10-2)}{1*(10-2)^2(10-4)} \approx 4.6875
D(X2)=1∗(10−2)2(10−4)2∗102(1+10−2)≈4.6875
所以
E
X
4
≈
6.25
EX^4 \approx 6.25
EX4≈6.25
3. 设随机变量(X,Y)的概率密度函数为
f
X
Y
(
x
,
y
)
=
A
s
i
n
(
x
+
y
)
,
0
≤
x
≤
π
2
,
0
≤
y
≤
π
2
f_{XY}(x,y)=Asin(x+y),0 \leq x \leq \frac{\pi}{2},0\leq y\leq \frac{\pi}{2}
fXY(x,y)=Asin(x+y),0≤x≤2π,0≤y≤2π
- 求系数A
- 求X的均值 m X m_X mX
- 求X的方差 σ X 2 \sigma ^2_X σX2
- 求X和Y的相关系数
ρ
X
Y
\rho_{XY}
ρXY
解:
(1)
1 = ∫ − ∞ + ∞ ∫ − ∞ + ∞ f ( x , y ) d x d y = ∫ 0 π / 2 ∫ 0 π / 2 A s i n ( x + y ) d x d y 1 = \int_{-\infty}^{+\infty}\int_{-\infty}^{+\infty}f(x,y)dxdy = \int_{0}^{\pi /2}\int_{0}^{\pi /2}Asin(x+y)dxdy 1=∫−∞+∞∫−∞+∞f(x,y)dxdy=∫0π/2∫0π/2Asin(x+y)dxdy
= ∫ 0 π / 2 ( ∫ 0 π / 2 A s i n ( x + y ) d x ) d y = \int_{0}^{\pi /2}(\int_{0}^{\pi /2}Asin(x+y)dx)dy =∫0π/2(∫0π/2Asin(x+y)dx)dy
= − ∫ 0 π / 2 ( A c o s ( x + y ) d x ) ∣ 0 π / 2 d y =- \int_{0}^{\pi /2}(Acos(x+y)dx)|_0^{\pi /2}dy =−∫0π/2(Acos(x+y)dx)∣0π/2dy
= ∫ 0 π / 2 ( − A c o s ( π / 2 + y ) + A c o s y ) d y = \int_{0}^{\pi /2}(-Acos(\pi /2+y)+Acosy)dy =∫0π/2(−Acos(π/2+y)+Acosy)dy
= 2 A = 2A =2A
所以A=1/2
(2)第一步,求出X的概率密度函数
f X ( x ) = ∫ 0 π / 2 1 / 2 s i n ( x + y ) d y = − 1 / 2 [ c o s ( π 2 + x ) − c o s x ] f_X(x)=\int_0^{\pi/2} 1/2sin(x+y)dy\\=-1/2[cos(\frac{\pi}{2}+x)-cosx] fX(x)=∫0π/21/2sin(x+y)dy=−1/2[cos(2π+x)−cosx]
根据公式求均值
m x = E x = − 1 / 2 ∫ 0 π 2 x c o s ( π 2 + x ) − x c o s x d x = π / 4 mx = Ex =-1/2 \int_0^{\frac{\pi}{2}}xcos(\frac{\pi}{2}+x)-xcosx dx\\ = \pi /4 mx=Ex=−1/2∫02πxcos(2π+x)−xcosxdx=π/4
(3)根据公式求解
X
2
的
期
望
X^2的期望
X2的期望,代入第二问中的
f
X
(
x
)
f_X(x)
fX(x)
E
x
2
=
∫
0
π
/
2
x
2
f
X
(
x
)
d
x
=
1
/
2
∫
0
π
/
2
x
2
(
s
i
n
x
+
c
o
s
x
)
d
x
=
π
2
+
π
2
8
−
2
Ex^2 = \int_0^{\pi /2}x^2f_X(x)dx\\= 1/2 \int_0^{\pi /2}x^2(sinx+cosx )dx\\=\frac{\pi}{2}+\frac{\pi ^2}{8}-2
Ex2=∫0π/2x2fX(x)dx=1/2∫0π/2x2(sinx+cosx)dx=2π+8π2−2
σ
X
2
=
D
X
=
E
x
2
−
(
E
x
)
2
=
π
2
+
π
2
8
−
2
−
(
π
/
4
)
2
\sigma ^2_X = DX = Ex^2-(Ex)^2\\ = \frac{\pi}{2}+\frac{\pi ^2}{8}-2 - (\pi /4)^2
σX2=DX=Ex2−(Ex)2=2π+8π2−2−(π/4)2
(4)求解相关系数的公式如下
ρ
X
Y
=
c
o
v
(
X
,
Y
)
D
X
D
Y
=
E
X
Y
−
E
X
E
Y
D
X
D
Y
\rho _{XY} =\frac{cov(X,Y)}{\sqrt{D_X} \sqrt{D_Y}} = \frac{E_{XY}-E_XE_Y}{\sqrt{D_X} \sqrt{D_Y}}
ρXY=DXDYcov(X,Y)=DXDYEXY−EXEY
所以现在需要求出
E
X
Y
E_{XY}
EXY
又
E
X
Y
=
∫
0
π
/
2
∫
0
π
/
2
1
/
2
x
y
s
i
n
(
x
+
y
)
d
x
d
y
E_{XY} = \int_0^{\pi/2}\int_0^{\pi/2} 1/2xysin(x+y)dxdy
EXY=∫0π/2∫0π/21/2xysin(x+y)dxdy
求解得到
E
X
Y
=
π
2
−
1
E_{XY} = \frac{\pi}{2}-1
EXY=2π−1
又根据第三问的结果,知道
D
X
=
σ
X
2
DX = \sigma ^2_X
DX=σX2
因为X,Y对称,所以
D
X
=
D
Y
D_X=D_Y
DX=DY
所以把
D
X
,
D
Y
,
E
X
Y
代
入
D_X,D_Y,E_{XY}代入
DX,DY,EXY代入得到相关系数
ρ
X
Y
=
E
X
Y
−
E
X
E
Y
D
X
D
Y
\rho _{XY} = \frac{E_{XY}-E_XE_Y}{\sqrt{D_X} \sqrt{D_Y}}
ρXY=DXDYEXY−EXEY
4. 设(X,Y)的联合概率密度函数为
f
(
x
,
y
)
=
{
21
4
x
2
y
,
x
2
≤
y
≤
1
0
,
o
t
h
e
r
s
w
i
s
e
f(x,y)=\begin{cases} \frac{21}{4}x^2y, &x^2 \leq y \leq 1\\ 0 ,& otherswise \end{cases}
f(x,y)={421x2y,0,x2≤y≤1otherswise
- 求给定条件Y=y(0<y<1)的条件下, X 2 X^2 X2的期望 E ( X 2 ∣ Y = 1 ) E(X^2|Y=1) E(X2∣Y=1)
- 求方差 D ( X 2 ∣ Y = 1 ) D(X^2|Y=1) D(X2∣Y=1)
- 设随机变量X的密度函数
f
(
x
)
=
∣
x
∣
,
∣
x
∣
<
1
f(x)=|x|,|x|<1
f(x)=∣x∣,∣x∣<1,求X的特征函数g(1)
解:
(1)
当 y ≤ 1 时 f Y ( y ) = ∫ − ∞ + ∞ f ( x , y ) d x = ∫ − 1 1 21 4 x 2 y d x = 7 2 y 当y\leq 1 时f_{Y(y)} = \int_{-\infty}^{+\infty}f(x,y)dx= \int_{-1}^{1}\frac{21}{4}x^2ydx =\frac{7}{2}y 当y≤1时fY(y)=∫−∞+∞f(x,y)dx=∫−11421x2ydx=27y
所 以 , 当 y ≤ 1 时 f ( x ∣ y ) = f ( x , y ) f Y ( y ) = 3 2 x 2 所以,当y\leq 1时f(x|y) =\frac{f(x,y)}{f_{Y(y)}} = \frac{3}{2}x^2 所以,当y≤1时f(x∣y)=fY(y)f(x,y)=23x2
E ( X 2 ∣ Y = y ) = ∫ − ∞ + ∞ x 2 f ( x ∣ y ) d x = ∫ − 1 1 3 2 x 4 d x = 3 5 E(X^2|Y=y) =\int_{-\infty}^{+\infty}x^2f(x|y)dx = \int_{-1}^{1}\frac{3}{2}x^4dx =\frac{3}{5} E(X2∣Y=y)=∫−∞+∞x2f(x∣y)dx=∫−1123x4dx=53
(2)因为根据公式
E
(
X
4
∣
Y
=
y
)
=
∫
−
1
1
x
4
f
(
x
∣
y
)
d
x
=
∫
−
1
1
x
4
∗
3
2
x
2
d
x
=
3
/
7
E(X^4|Y=y) = \int_{-1}^{1}x^4f(x|y)dx\\= \int_{-1}^{1}x^4*\frac{3}{2}x^2dx = 3/7
E(X4∣Y=y)=∫−11x4f(x∣y)dx=∫−11x4∗23x2dx=3/7
所以
D
X
2
=
E
X
4
−
(
E
X
)
2
=
3
/
7
−
(
3
/
5
)
2
=
0.0686
DX^2 = EX^4-(EX)^2\\=3/7-(3/5)^2 = 0.0686
DX2=EX4−(EX)2=3/7−(3/5)2=0.0686
(3)连续性随机变量的特征函数公式
g
(
t
)
=
∫
−
∞
+
∞
e
i
t
x
f
(
x
)
d
x
g(t) = \int_{-\infty}^{+\infty}e^{itx}f(x)dx
g(t)=∫−∞+∞eitxf(x)dx
g
(
1
)
=
∫
−
1
1
e
i
x
∣
x
∣
d
x
g(1) = \int_{-1}^{1}e^{ix}|x|dx
g(1)=∫−11eix∣x∣dx
根据欧拉公式
e
(
i
t
x
)
=
c
o
s
t
(
x
)
+
i
s
i
n
(
t
x
)
e^{(itx)} = cost(x)+isin(tx)
e(itx)=cost(x)+isin(tx)
∫
−
1
1
e
i
x
∣
x
∣
d
x
=
∫
−
1
1
∣
x
∣
(
c
o
s
x
+
i
s
i
n
x
)
d
x
\int_{-1}^{1} e^{ix} |x| dx =\int_{-1}^{1}|x|( cosx+isinx )dx
∫−11eix∣x∣dx=∫−11∣x∣(cosx+isinx)dx
因为|x| isinx 是奇函数,所以
∫
−
1
1
∣
x
∣
i
s
i
n
x
d
x
=
0
\int_{-1}^{1}|x| isinx dx = 0
∫−11∣x∣isinxdx=0,则
g
(
1
)
=
∫
−
1
1
∣
x
∣
c
o
s
x
d
x
=
∫
−
1
0
x
c
o
s
x
d
x
+
∫
0
1
x
c
o
s
x
d
x
g(1) = \int_{-1}^{1}|x| cosxdx = \int_{-1}^{0}x cosxdx +\int_{0}^{1}xcosxdx
g(1)=∫−11∣x∣cosxdx=∫−10xcosxdx+∫01xcosxdx
=
2
(
c
o
s
1
+
s
i
n
1
−
1
)
=
0.7635
= 2(cos1+sin1-1) = 0.7635
=2(cos1+sin1−1)=0.7635