1 事件关系与运算
1.1 事件关系
集合(事件)关系 | 集合语言 | 语言描述 | 概率关系 |
---|---|---|---|
包含 | A ⊆ B A\subseteq B A⊆B | A A A 发生,则 B B B 一定发生 | P ( A ) ⩽ P ( B ) P(A)\leqslant P(B) P(A)⩽P(B) |
等价 | A = B A=B A=B | A A A 发生 B B B 一定发生,反之亦然 | P ( A ) = P ( B ) P(A)=P(B) P(A)=P(B) |
交 | ⋂ k = 1 n A k \displaystyle\bigcap_{k=1}^{n}A_k k=1⋂nAk | 事件 A k A_k Ak 同时发生 | A k A_k Ak 互相独立, P ( ⋂ k = 1 n A k ) = ∏ k = 1 n P ( A k ) \displaystyle P\left(\bigcap_{k=1}^{n}A_k\right)=\prod_{k=1}^{n}P(A_k) P(k=1⋂nAk)=k=1∏nP(Ak) |
并 | ⋃ k = 1 n A k \displaystyle\bigcup_{k=1}^{n}A_k k=1⋃nAk | A k A_k Ak 至少有一个发生 | ∀ i ≠ j , A i A j = ϕ , P ( ⋃ k = 1 n A k ) = ∑ k = 1 n P ( A k ) \forall i\ne j,A_iA_j=\phi,\\P\left(\displaystyle\bigcup_{k=1}^{n}A_k\right)=\sum\limits_{k=1}^{n}P(A_k) ∀i=j,AiAj=ϕ,P(k=1⋃nAk)=k=1∑nP(Ak) |
减 | A − B = A B ˉ A-B=A\bar{B} A−B=ABˉ | A − B A-B A−B 发生,仅当 A A A 发生 B B B 不发生 | P ( A − B ) = P ( A B ˉ ) = P ( A ) − P ( A B ) \begin{aligned}P(A-B)&=P(A\bar{B})\\&=P(A)-P(AB)\end{aligned} P(A−B)=P(ABˉ)=P(A)−P(AB) |
对立 | A ˉ = Ω − A \bar{A}=\Omega-A Aˉ=Ω−A | A ˉ \bar{A} Aˉ 发生,当且仅当 A A A 不发生 | P ( A ˉ ) = P ( Ω ) − P ( A ) = 1 − P ( A ) \begin{aligned}P(\bar{A})&=P(\Omega)-P(A)\\&=1-P(A)\end{aligned} P(Aˉ)=P(Ω)−P(A)=1−P(A) |
互斥 | A ∩ B = ϕ A\cap B=\phi A∩B=ϕ | A , B A,B A,B 不可能同时发生 | P ( A B ) = P ( ϕ ) = 0 P(AB)=P(\phi)=0 P(AB)=P(ϕ)=0 |
完备事件组 | ⋃ k = 1 n A k = Ω , A i ∩ A j = ϕ ( 1 ⩽ i < j ⩽ n ) \displaystyle\bigcup_{k=1}^{n}A_k=\Omega,\\ \underset{(1\leqslant i<j\leqslant n)}{A_i\cap A_j=\phi} k=1⋃nAk=Ω,(1⩽i<j⩽n)Ai∩Aj=ϕ | 一个两两互斥的事件组,它们的和等于必然事件 | P ( ⋃ k = 1 n A k ) = P ( Ω ) = 1 P\left(\bigcup\limits_{k=1}^{n}A_k\right)=P(\Omega)=1 P(k=1⋃nAk)=P(Ω)=1 |
1.2 运算性质
运算律 | 集合语言(并) | 集合语言(交) |
---|---|---|
交换律 | A ∪ B = B ∪ A A\cup B=B\cup A A∪B=B∪A | A B = B A AB=BA AB=BA |
结合律 | ( A ∪ B ) ∪ C = A ∪ ( B ∪ C ) (A\cup B)\cup C=A\cup(B\cup C) (A∪B)∪C=A∪(B∪C) | ( A B ) C = A ( B C ) (AB)C=A(BC) (AB)C=A(BC) |
分配律 | ( A ∪ B ) ∩ C = A C ∪ B C (A\cup B)\cap C=AC\cup BC (A∪B)∩C=AC∪BC | ( A ∩ B ) ∪ C = ( A ∪ C ) ∩ ( B ∪ C ) (A\cap B)\cup C=(A\cup C)\cap (B\cup C) (A∩B)∪C=(A∪C)∩(B∪C) |
对偶律(德摩根律) | ⋃ k = 1 n A k ˉ = ⋂ k = 1 n A k ˉ \displaystyle\bar{\bigcup_{k=1}^{n}A_k}=\bigcap_{k=1}^{n}\bar{A_k} k=1⋃nAkˉ=k=1⋂nAkˉ | ⋃ k = 1 n A k ˉ = ⋂ k = 1 n A k ˉ \displaystyle\bigcup_{k=1}^{n}\bar{A_k}=\bar{\bigcap_{k=1}^{n}A_k} k=1⋃nAkˉ=k=1⋂nAkˉ |
2 概率运算
2.1 性质
2.1.1 不可能事件与必然事件
不可能事件:
P
(
ϕ
)
=
0
P(\phi)=0
P(ϕ)=0
必然事件:
P
(
Ω
)
=
1
P(\Omega)=1
P(Ω)=1
Tips:
不可能事件概率为0,但概率为零的未必是不可能事件;概率为0未必是空集,概率为1未必是全集。
2.1.2 概率关系与事件关系
事件关系可推出概率关系,但概率关系无法推知事件关系。
点与线的面积为零,概率为零,例如:
P ( A ) = 0 ⇏ ⇐ A = ϕ P(A)=0 \begin{array}{c} \nRightarrow\\ \Leftarrow\\ \end{array} A=\phi P(A)=0⇏⇐A=ϕ
P ( A B ) = 0 ⇏ ⇐ A B = ϕ P(AB)=0 \begin{array}{c} \nRightarrow\\ \Leftarrow\\ \end{array} AB=\phi P(AB)=0⇏⇐AB=ϕ
P ( A ) = P ( A B ) ⇏ A = A B ⊂ B P(A)=P(AB)\nRightarrow A=AB\subset B P(A)=P(AB)⇏A=AB⊂B
2.2 公式
2.2.1 加法公式
P ( A ∪ B ) = P ( A ) + P ( B ) − P ( A B ) = A , B 互斥 ⇔ P ( A B ) = 0 P ( A ) + P ( B ) \quad P(A\cup B)=P(A)+P(B)-P(AB)\xlongequal{A,B互斥\Leftrightarrow P(AB)=0}P(A)+P(B) P(A∪B)=P(A)+P(B)−P(AB)A,B互斥⇔P(AB)=0P(A)+P(B)
推论:
P
(
A
)
+
P
(
B
)
−
1
⩽
P
(
A
∪
B
)
⩽
1
P
(
A
B
)
=
P
(
A
)
+
P
(
B
)
−
P
(
A
∪
B
)
⩽
0
⩽
P
(
A
∪
B
)
P
(
A
)
+
P
(
B
)
\begin{aligned} P(A)+P(B)-1 &\overset{_{P(A\cup B)\leqslant 1}}{\leqslant} P(AB)=P(A)+P(B)-P(A\cup B) \\ &\overset{_{0\leqslant P(A\cup B)}}{\leqslant}P(A)+P(B) \end{aligned}
P(A)+P(B)−1⩽P(A∪B)⩽1P(AB)=P(A)+P(B)−P(A∪B)⩽0⩽P(A∪B)P(A)+P(B)
有限可加性:
A i A_i Ai 两两互斥 (即 ∀ i ≠ j , A i A j = ϕ \forall\ i\ne j,A_iA_j=\phi ∀ i=j,AiAj=ϕ),则 P ( ⋃ k = 1 n A k ) = ∑ k = 1 n P ( A k ) P\left (\bigcup\limits_{k=1}^{n}A_k\right)=\sum\limits_{k=1}^{n}P(A_k) P(k=1⋃nAk)=k=1∑nP(Ak)
2.2.2 减法公式
P ( A − B ) = P ( A B ˉ ) = P ( A − A B ) = P ( A ) − P ( A B ) P(A-B)=P(A\bar{B})=P(A-AB)=P(A)-P(AB) P(A−B)=P(ABˉ)=P(A−AB)=P(A)−P(AB)
推论:
P
(
A
−
B
)
=
P
(
A
)
−
P
(
A
B
)
⩾
A
B
⊆
B
⇒
P
(
A
B
)
⩽
P
(
B
)
P
(
A
)
−
P
(
B
)
P(A-B)=P(A)-P(AB)\overset{_{AB\subseteq B\Rightarrow P(AB)\leqslant P(B)}}{\geqslant}P(A)-P(B)
P(A−B)=P(A)−P(AB)⩾AB⊆B⇒P(AB)⩽P(B)P(A)−P(B)
2.2.3 对立事件
P ( A ˉ ) = 1 − P ( A ) , P ( A ˉ ∣ B ) = 1 − P ( A ∣ B ) P(\bar{A})=1-P(A)\ ,P(\bar{A}|B)=1-P(A|B) P(Aˉ)=1−P(A) ,P(Aˉ∣B)=1−P(A∣B)
2.2.4 条件概率
P ( A ∣ B ) = P ( A B ) P ( B ) = A , B 独立 ⇔ P ( A B ) = P ( A ) P ( B ) P ( A ) P(A|B)=\frac{P(AB)}{P(B)} \xlongequal{A,B\ 独立\Leftrightarrow P(AB)=P(A)P(B)}P(A) P(A∣B)=P(B)P(AB)A,B 独立⇔P(AB)=P(A)P(B)P(A)
意义:
事件A<在事件B发生的条件下>发生的概率;条件概率的作用为限制条件,缩减样本空间,故所有运算法则皆可在“|”前使用。
2.2.5 乘法公式
P ( A B ) = { P ( A ) P ( B ∣ A ) , P ( A ) > 0 P ( B ) P ( A ∣ B ) , P ( B ) > 0 P(AB)=\left\{ \begin{array}{l} P(A)P(B|A)\ ,P(A)>0\\ \\ P(B)P(A|B)\ ,P(B)>0\\ \end{array} \right. P(AB)=⎩ ⎨ ⎧P(A)P(B∣A) ,P(A)>0P(B)P(A∣B) ,P(B)>0
推广:
P
(
∏
k
=
1
n
A
k
)
=
∏
k
=
1
n
P
(
A
k
∣
⋂
i
=
0
k
=
1
A
i
)
,
A
0
=
ϕ
,
P
(
⋂
k
=
1
n
A
k
)
>
0
P\left(\prod_{k=1}^{n}A_k\right)=\prod_{k=1}^{n}P\left(A_k|\bigcap_{i=0}^{k=1}A_i\right),{\color{brown}A_0=\phi,P\left(\bigcap_{k=1}^{n}A_k\right)>0}
P(k=1∏nAk)=k=1∏nP(Ak∣i=0⋂k=1Ai),A0=ϕ,P(k=1⋂nAk)>0
推论:
P
(
A
B
)
=
0
⇒
P
(
A
B
C
)
=
P
(
A
B
)
P
(
A
B
C
∣
A
B
)
=
0
P(AB)=0\ \Rightarrow\ P(ABC)=P(AB)P(ABC|AB)=0
P(AB)=0 ⇒ P(ABC)=P(AB)P(ABC∣AB)=0
2.2.6 全概率公式
设 A i A_i Ai 为完备事件组,则有
P ( B ) = P ( B Ω ) = P ( B ⋃ k = 1 n A k ) = ∑ k = 1 n P ( B A k ) = P ( A i ) > 0 ∑ k = 1 n P ( A k ) P ( B ∣ A k ) P(B)=P(B\Omega)=P\left(B\bigcup_{k=1}^{n}A_k\right)=\sum_{k=1}^n P(BA_k)\xlongequal{P(A_i)>0}\sum_{k=1}^{n}P(A_k)P(B|A_k) P(B)=P(BΩ)=P(Bk=1⋃nAk)=k=1∑nP(BAk)P(Ai)>0k=1∑nP(Ak)P(B∣Ak)
意义:
随机试验分为两个阶段: 1 → P ( A i ) A i → P ( B ∣ A i ) B 1\xrightarrow{P(A_i)}A_i\xrightarrow{P(B|A_i)}B 1P(Ai)AiP(B∣Ai)B,第一阶段多种可能结果为 A i A_i Ai,求解第二阶段某事件 B B B 的概率.
2.2.7 贝叶斯公式
设 A i A_i Ai 为完备事件组,则有
∀ 1 ⩽ m ⩽ n : P ( A m ∣ B ) = P ( B ) > 0 P ( B A m ) P ( B ) = P ( A i ) > 0 P ( A m ) P ( B ∣ A m ) ∑ k = 1 n P ( A k ) P ( B ∣ A k ) \forall 1\leqslant m\leqslant n:P(A_m|B)\xlongequal{P(B)>0}\frac{P(BA_m)}{P(B)}\xlongequal{P(A_i)>0}\frac{P(A_m)P(B|A_m)}{\displaystyle\sum_{k=1}^{n}P(A_k)P(B|A_k)} ∀1⩽m⩽n:P(Am∣B)P(B)>0P(B)P(BAm)P(Ai)>0k=1∑nP(Ak)P(B∣Ak)P(Am)P(B∣Am)
2.2.8 常见概率不等关系
{ ∀ A : 0 ⩽ P ( A ) ⩽ 1 A ⊆ B , P ( A ) ⩽ P ( B ) A B ⊆ A , P ( A B ) ⩽ P ( A ) ⇒ { 0 ⩽ P ( A B C ) ⩽ P ( A B ) ⩽ P ( A ) P ( B ) ⩽ P ( A ∪ B ) ⩽ P ( A ∪ B ∪ C ) ⩽ 1 P ( A ) + P ( B ) − 1 ⩽ P ( A B ) = P ( A ) + P ( B ) − P ( A ∪ B ) ⩽ P ( A ) + P ( B ) P ( A − B ) = P ( A B ˉ ) = P ( A − A B ) = P ( A ) − P ( A B ) ⩾ P ( A ) − P ( B ) \begin{aligned} &\left\{ \begin{array}{l} \forall A:0\leqslant P(A)\leqslant 1\\ A\subseteq B\ ,P(A)\leqslant P(B)\\ AB\subseteq A\ ,P(AB)\leqslant P(A)\\ \end{array} \right. \\ \Rightarrow &\left\{ \begin{array}{l} 0\leqslant P(ABC)\leqslant P(AB)\leqslant P(A)\\ \\ P(B)\leqslant P(A\cup B)\leqslant P(A\cup B\cup C)\leqslant 1\\ \\ P(A)+P(B)-1\leqslant P(AB)=P(A)+P(B)-P(A\cup B)\leqslant P(A)+P(B)\\ \\ P(A-B)=P(A\bar{B})=P(A-AB)=P(A)-P(AB)\geqslant P(A)-P(B)\\ \end{array} \right. \end{aligned} ⇒⎩ ⎨ ⎧∀A:0⩽P(A)⩽1A⊆B ,P(A)⩽P(B)AB⊆A ,P(AB)⩽P(A)⎩ ⎨ ⎧0⩽P(ABC)⩽P(AB)⩽P(A)P(B)⩽P(A∪B)⩽P(A∪B∪C)⩽1P(A)+P(B)−1⩽P(AB)=P(A)+P(B)−P(A∪B)⩽P(A)+P(B)P(A−B)=P(ABˉ)=P(A−AB)=P(A)−P(AB)⩾P(A)−P(B)
Tips: 事件交更小,并更大
2.3 独立
2.3.1 相互独立
A , B 相互独立 ⇔ d e f P ( A B ) = P ( A ) P ( B ) ⇔ A , B ˉ 独立 ⇔ A ˉ , B 独立 ⇔ A ˉ , B ˉ 独立 ⇔ P ( A ∣ B ) = P ( A ∣ B ˉ ) ( 0 < P ( B ) < 1 ) ⇔ P ( A ˉ ∣ B ) + P ( A ∣ B ˉ ) = 1 ( 0 < P ( B ) < 1 ) ⇔ P ( A ∣ B ) + P ( A ˉ ∣ B ˉ ) = 1 ( 0 < P ( B ) < 1 ) \begin{aligned} A,B 相互独立 &\xLeftrightarrow{\mathrm{\ def\ }}\ P(AB)=P(A)P(B) \\&\Leftrightarrow A,\bar{B}独立 \Leftrightarrow \bar{A},B 独立 \Leftrightarrow \bar{A},\bar{B} 独立 \\&\Leftrightarrow P(A|B)=P(A|\bar{B})\ (0<P(B)<1) \\&\Leftrightarrow P(\bar{A}|B)+P(A|\bar{B})=1\ (0<P(B)<1) \\&\Leftrightarrow P(A|B)+P(\bar{A}|\bar{B})=1\ (0<P(B)<1) \end{aligned} A,B相互独立 def P(AB)=P(A)P(B)⇔A,Bˉ独立⇔Aˉ,B独立⇔Aˉ,Bˉ独立⇔P(A∣B)=P(A∣Bˉ) (0<P(B)<1)⇔P(Aˉ∣B)+P(A∣Bˉ)=1 (0<P(B)<1)⇔P(A∣B)+P(Aˉ∣Bˉ)=1 (0<P(B)<1)
A , B , C 相互独立 ⇔ { P ( A B ) = P ( A ) P ( B ) P ( B C ) = P ( B ) P ( C ) P ( C A ) = P ( C ) P ( A ) } ⇔ A , B , C 两两独立 P ( A B C ) = P ( A ) P ( B ) P ( C ) A,B,C相互独立 \Leftrightarrow \left\{\begin{array}{l} \left.\begin{array}{r} P(AB)=P(A)P(B)\\ P(BC)=P(B)P(C)\\ P(CA)=P(C)P(A)\\ \end{array} \right\} \Leftrightarrow A,B,C两两独立\\ P(ABC)=P(A)P(B)P(C)\\ \end{array}\right. A,B,C相互独立⇔⎩ ⎨ ⎧P(AB)=P(A)P(B)P(BC)=P(B)P(C)P(CA)=P(C)P(A)⎭ ⎬ ⎫⇔A,B,C两两独立P(ABC)=P(A)P(B)P(C)
2.3.2 独立与互斥
-
A , B A,B A,B 独立 ⇔ P ( A B ) = P ( A ) P ( B ) ⇎ A , B \Leftrightarrow P(AB)=P(A)P(B) \nLeftrightarrow A,B ⇔P(AB)=P(A)P(B)⇎A,B互斥 ⇔ A , B = ϕ ⇒ P ( A B ) = 0 \Leftrightarrow A,B=\phi \Rightarrow P(AB)=0 ⇔A,B=ϕ⇒P(AB)=0
-
P ( A ) , P ( B ) > 0 P(A),P(B)>0 P(A),P(B)>0
互斥一定不独立: P ( A B ) = P ( ϕ ) = 0 ≠ P ( A ) P ( B ) > 0 P(AB)=P(\phi)=0\ne P(A)P(B)>0 P(AB)=P(ϕ)=0=P(A)P(B)>0
独立一定不互斥: P ( A B ) = P ( A ) P ( B ) > 0 ⇏ P ( A B ) = 0 P(AB)=P(A)P(B)>0\nRightarrow P(AB)=0 P(AB)=P(A)P(B)>0⇏P(AB)=0 -
P ( A ) = 0 , A P(A)=0,A P(A)=0,A 与 ∀ B \forall B ∀B 独立; A , B A,B A,B 也可能互斥,如 P ( A ) P ( B ) = 0 P(A)P(B)=0 P(A)P(B)=0
-
0 < P ( A ) , P ( B ) < 1 0<P(A),P(B)<1 0<P(A),P(B)<1,且 A B = ϕ AB=\phi AB=ϕ 或 A ⊂ B ⇒ A , B A\subset B\Rightarrow A,B A⊂B⇒A,B 不独立.
-
ϕ \phi ϕ 与任意事件既独立又互斥.
3 常见一维二维分布与三大抽样分布
3.1 一维r.v.
3.1.1 离散型r.v.
3.1.1.1 0-1分布 X ∼ B ( 1 , p ) X\thicksim B(1,p) X∼B(1,p)
意义:
1 1 1 次伯努利试验,成功率为 p p p,成功次数 X X X 服从0-1分布.
概率分布:
P { X = k } = P k ( 1 − p ) 1 − k ( 0 < p < 1 ; k = 0 , 1 ) P\{X=k\}=P^k(1-p)^{1-k}\ (0<p<1;k=0,1) P{X=k}=Pk(1−p)1−k (0<p<1;k=0,1)
数字特征:
E ( X ) = p , D ( X ) = p ( 1 − p ) E(X)=p\ ,\ D(X)=p(1-p) E(X)=p , D(X)=p(1−p)
proof:
E ( X ) = 0 ⋅ P { X = 0 } + 1 ⋅ P { X = 1 } = 0 ⋅ ( 1 − p ) + 1 ⋅ p = p E ( X 2 ) = 0 2 ⋅ P { X 2 = 0 } + 1 2 ⋅ P { X 2 = 1 } = p D ( X ) = E ( X 2 ) − [ E ( X ) ] 2 = p ( 1 − p ) \begin{aligned} &E(X)=0\cdot P\{X=0\}+1\cdot P\{X=1\}=0\cdot (1-p)+1\cdot p=p \\&E(X^2)=0^2\cdot P\{X^2=0\}+1^2\cdot P\{X^2=1\}=p \\&D(X)=E(X^2)-[E(X)]^2=p(1-p) \end{aligned} E(X)=0⋅P{X=0}+1⋅P{X=1}=0⋅(1−p)+1⋅p=pE(X2)=02⋅P{X2=0}+12⋅P{X2=1}=pD(X)=E(X2)−[E(X)]2=p(1−p)
3.1.1.2 二项分布 X ∼ B ( n , p ) X\thicksim B(n,p) X∼B(n,p)
意义:
n n n 次独立重复伯努利试验,成功率为 p p p,成功次数 X X X 服从二项分布.
概率分布:
P { X = k } = C n k p k ( 1 − p ) n − k ( 0 < p < 1 ; k = 0 , 1 , ⋯ , n ) P\{X=k\}=C_n^k\ p^k(1-p)^{n-k}\ (0<p<1;k=0,1,\cdots,n) P{X=k}=Cnk pk(1−p)n−k (0<p<1;k=0,1,⋯,n)
独立可加性:
{ X ∼ B ( n , p ) Y ∼ B ( m , p ) ⇒ Z = X + Y ∼ B ( m + n , p ) \left\{\begin{array}{l}X\sim B(n,p)\\Y\sim B(m,p)\\ \end{array}\right. \Rightarrow Z=X+Y\sim B(m+n,p) {X∼B(n,p)Y∼B(m,p)⇒Z=X+Y∼B(m+n,p)
数字特征:
E ( X ) = n p , D ( X ) = n p ( 1 − p ) E(X)=np\ ,\ D(X)=np(1-p) E(X)=np , D(X)=np(1−p)
proof1:
将
X
X
X 分解为
n
n
n 个独立同分布于
0
−
1
0-1
0−1 分布的随机变量,即
X
=
∑
k
=
1
n
X
k
∼
B
(
n
,
p
)
,
X
k
∼
B
(
1
,
p
)
X=\sum_{k=1}^nX_k\sim B(n,p)\ ,\ X_k\sim B(1,p)
X=k=1∑nXk∼B(n,p) , Xk∼B(1,p)
由 0-1 分布有
E
(
X
k
)
=
E
(
X
k
2
)
=
p
,
D
(
X
k
)
=
E
(
X
k
2
)
−
E
(
X
k
)
2
=
p
(
1
−
p
)
E(X_k)=E(X_k^2)=p\ ,\ D(X_k)=E(X_k^2)-E(X_k)^2=p(1-p)
E(Xk)=E(Xk2)=p , D(Xk)=E(Xk2)−E(Xk)2=p(1−p)
于是
E
(
X
)
=
∑
k
=
1
n
E
(
X
k
)
=
n
p
,
D
(
X
)
=
∑
k
=
1
n
D
(
X
k
)
=
n
p
(
1
−
p
)
E(X)=\sum_{k=1}^nE(X_k)=np\ ,D(X)=\sum_{k=1}^nD(X_k)=np(1-p)
E(X)=k=1∑nE(Xk)=np ,D(X)=k=1∑nD(Xk)=np(1−p)
proof2:
E
(
X
)
=
∑
k
=
0
n
k
⋅
P
{
X
=
k
}
=
∑
k
=
0
n
k
⋅
C
n
k
p
k
(
1
−
p
)
n
−
k
=
∑
k
=
1
n
k
⋅
C
n
k
p
k
(
1
−
p
)
n
−
k
=
∑
k
=
1
n
k
⋅
n
!
k
!
(
n
−
k
)
!
p
k
(
1
−
p
)
n
−
k
=
n
p
∑
k
=
1
n
(
n
−
1
)
!
(
k
−
1
)
!
(
n
−
k
)
!
p
k
−
1
(
1
−
p
)
(
n
−
1
)
−
(
k
−
1
)
=
i
=
k
−
1
n
p
∑
i
=
0
n
−
1
(
n
−
1
)
!
i
!
[
(
n
−
1
)
−
i
]
!
p
i
(
1
−
p
)
(
n
−
1
)
−
i
=
n
p
∑
i
=
0
n
−
1
C
n
−
1
i
p
i
(
1
−
p
)
(
n
−
1
)
−
i
=
n
p
[
p
+
(
1
−
p
)
]
n
−
1
=
n
p
E
(
X
2
)
=
∑
k
=
0
n
k
2
⋅
C
n
k
p
k
(
1
−
p
)
n
−
k
=
∑
k
=
0
n
[
k
(
k
−
1
)
+
k
]
C
n
k
p
k
(
1
−
p
)
n
−
k
=
∑
k
=
0
n
k
(
k
−
1
)
C
n
k
p
k
(
1
−
p
)
n
−
k
+
E
(
X
)
=
∑
k
=
0
n
k
(
k
−
1
)
n
!
k
!
(
n
−
k
)
!
p
k
(
1
−
p
)
n
−
k
+
n
p
=
n
(
n
−
1
)
p
2
∑
k
=
0
n
(
n
−
2
)
!
(
k
−
2
)
!
(
n
−
k
)
!
p
k
−
2
(
1
−
p
)
(
n
−
2
)
−
(
k
−
2
)
+
n
p
=
n
(
n
−
1
)
p
2
[
p
+
(
1
−
p
)
]
n
−
2
+
n
p
=
n
(
n
−
1
)
p
2
+
n
p
D
(
X
)
=
E
(
X
2
)
−
[
E
(
X
)
]
2
=
n
p
(
1
−
p
)
\begin{aligned} E(X)&=\sum_{k=0}^{n}k\cdot P\{X=k\} =\sum_{k=0}^{n}k\cdot C_n^kp^k(1-p)^{n-k} \\&=\sum_{k=1}^{n}k\cdot C_n^kp^k(1-p)^{n-k} =\sum_{k=1}^{n}k\cdot\frac{n!}{k!(n-k)!}p^k(1-p)^{n-k} \\&=np\sum_{k=1}^{n}\frac{(n-1)!}{(k-1)!(n-k)!}p^{k-1}(1-p)^{(n-1)-(k-1)} \\&\xlongequal{i=k-1}np\sum_{i=0}^{n-1}\frac{(n-1)!}{i![(n-1)-i]!}p^i(1-p)^{(n-1)-i} \\&=np\sum_{i=0}^{n-1}C_{n-1}^{i}p^i(1-p)^{(n-1)-i} =np[p+(1-p)]^{n-1}=np \\ \\E(X^2)&=\sum_{k=0}^{n}k^2\cdot C_n^kp^k(1-p)^{n-k} =\sum_{k=0}^{n}[k(k-1)+k]C_n^kp^k(1-p)^{n-k} \\&=\sum_{k=0}^{n}k(k-1)C_n^kp^k(1-p)^{n-k}+E(X) \\&=\sum_{k=0}^{n}k(k-1)\frac{n!}{k!(n-k)!}p^k(1-p)^{n-k}+np \\&=n(n-1)p^2\sum_{k=0}^{n}\frac{(n-2)!}{(k-2)!(n-k)!}p^{k-2}(1-p)^{(n-2)-(k-2)}+np \\&=n(n-1)p^2[p+(1-p)]^{n-2}+np=n(n-1)p^2+np \\ \\D(X)&=E(X^2)-[E(X)]^2=np(1-p) \end{aligned}
E(X)E(X2)D(X)=k=0∑nk⋅P{X=k}=k=0∑nk⋅Cnkpk(1−p)n−k=k=1∑nk⋅Cnkpk(1−p)n−k=k=1∑nk⋅k!(n−k)!n!pk(1−p)n−k=npk=1∑n(k−1)!(n−k)!(n−1)!pk−1(1−p)(n−1)−(k−1)i=k−1npi=0∑n−1i![(n−1)−i]!(n−1)!pi(1−p)(n−1)−i=npi=0∑n−1Cn−1ipi(1−p)(n−1)−i=np[p+(1−p)]n−1=np=k=0∑nk2⋅Cnkpk(1−p)n−k=k=0∑n[k(k−1)+k]Cnkpk(1−p)n−k=k=0∑nk(k−1)Cnkpk(1−p)n−k+E(X)=k=0∑nk(k−1)k!(n−k)!n!pk(1−p)n−k+np=n(n−1)p2k=0∑n(k−2)!(n−k)!(n−2)!pk−2(1−p)(n−2)−(k−2)+np=n(n−1)p2[p+(1−p)]n−2+np=n(n−1)p2+np=E(X2)−[E(X)]2=np(1−p)
3.1.1.3 几何分布 X ∼ G ( n , p ) X\thicksim G(n,p) X∼G(n,p)
意义:
n次独立重复伯努利试验,成功率为 p p p,首次成功次数 X X X 服从几何分布.
概率分布:
P { X = n } = p ( 1 − p ) n − 1 ( 0 < p < 1 ; n = 0 , 1 , ⋯ ) P\{X=n\}=p(1-p)^{n-1}\ (0<p<1;\ n=0,1,\cdots) P{X=n}=p(1−p)n−1 (0<p<1; n=0,1,⋯)
数字特征:
E ( X ) = 1 p , D ( X ) = 1 − p p 2 E(X)=\frac{1}{\ p\ }\ ,\ D(X)=\frac{1-p}{p^2} E(X)= p 1 , D(X)=p21−p
proof:
E
(
X
)
=
∑
n
=
0
+
∞
n
p
(
1
−
p
)
n
−
1
=
p
⋅
d
d
(
1
−
p
)
[
∑
n
=
0
+
∞
(
1
−
p
)
n
]
=
p
⋅
d
d
(
1
−
p
)
[
1
1
−
(
1
−
p
)
]
=
1
p
E
(
X
2
)
=
∑
n
=
0
+
∞
n
2
p
(
1
−
p
)
n
−
1
=
p
⋅
d
d
(
1
−
p
)
[
∑
n
=
0
+
∞
n
(
1
−
p
)
n
]
=
p
⋅
d
d
(
1
−
p
)
[
(
1
−
p
)
⋅
∑
n
=
0
+
∞
n
(
1
−
p
)
n
−
1
]
=
p
⋅
d
d
(
1
−
p
)
{
(
1
−
p
)
⋅
d
d
(
1
−
p
)
[
∑
n
=
0
+
∞
(
1
−
p
)
n
]
}
=
x
=
1
−
p
p
⋅
[
x
⋅
(
1
1
−
x
)
′
]
′
=
p
1
−
x
2
(
1
−
x
)
4
=
2
−
p
p
2
D
(
X
)
=
E
(
X
2
)
−
[
E
(
X
)
]
2
=
1
−
p
p
2
\begin{aligned} E(X)&=\sum_{n=0}^{+\infty}np(1-p)^{n-1}=p\cdot\frac{d}{\mathrm{d}(1-p)}\left[\sum_{n=0}^{+\infty}(1-p)^n\right] \\&=p\cdot\frac{d}{\mathrm{d}(1-p)}\left[\frac{1}{1-(1-p)}\right]=\frac{1}{p} \\E(X^2)&=\sum_{n=0}^{+\infty}n^2p(1-p)^{n-1}=p\cdot\frac{\mathrm{d}}{\mathrm{d}(1-p)}\left[\sum_{n=0}^{+\infty}n(1-p)^n\right] \\&=p\cdot\frac{\mathrm{d}}{\mathrm{d}(1-p)}\left[(1-p)\cdot\sum_{n=0}^{+\infty}n(1-p)^{n-1}\right] \\&=p\cdot\frac{\mathrm{d}}{\mathrm{d}(1-p)}\left\{(1-p)\cdot\frac{\mathrm{d}}{\mathrm{d}(1-p)}\left[\sum_{n=0}^{+\infty}(1-p)^{n}\right]\right\} \\&\xlongequal{x=1-p}p\cdot\left[x\cdot\left(\frac{1}{1-x}\right)'\right]'=p\frac{1-x^2}{(1-x)^4}=\frac{2-p}{p^2} \\D(X)&=E(X^2)-[E(X)]^2=\frac{1-p}{p^2} \end{aligned}
E(X)E(X2)D(X)=n=0∑+∞np(1−p)n−1=p⋅d(1−p)d[n=0∑+∞(1−p)n]=p⋅d(1−p)d[1−(1−p)1]=p1=n=0∑+∞n2p(1−p)n−1=p⋅d(1−p)d[n=0∑+∞n(1−p)n]=p⋅d(1−p)d[(1−p)⋅n=0∑+∞n(1−p)n−1]=p⋅d(1−p)d{(1−p)⋅d(1−p)d[n=0∑+∞(1−p)n]}x=1−pp⋅[x⋅(1−x1)′]′=p(1−x)41−x2=p22−p=E(X2)−[E(X)]2=p21−p
3.1.1.4 超几何分布 X ∼ H ( n , M , N ) X\thicksim H(n,M,N) X∼H(n,M,N)
意义:
N 件产品中 M 件不合格,随机抽 n 件,发现 k 件不合格的概率.
概率分布:
P { X = n } = C M k C N − M n − k C N n ( k = 1 , 2 , ⋯ , min { n , M } ; n , M ⩽ n ) P\{X=n\}=\frac{C_{_M}^kC_{_{N-M}}^{n-k}}{C_{_N}^n}\ (k=1,2,\cdots,\min\{n,M\};n,M\leqslant n) P{X=n}=CNnCMkCN−Mn−k (k=1,2,⋯,min{n,M};n,M⩽n)
数字特征:
E ( X ) = n M N , D ( X ) = n M N ( 1 − M N ) N − n N − 1 E(X)=\frac{nM}{N}\ ,\ D(X)=\frac{nM}{N}\left(1-\frac{M}{N}\right)\frac{N-n}{N-1} E(X)=NnM , D(X)=NnM(1−NM)N−1N−n
3.1.1.5 泊松分布 X ∼ P ( λ ) X\thicksim P(\lambda) X∼P(λ)
意义:
λ \lambda λ 为单位时间(或单位面积)内随机事件的平均发生次数.
概率分布:
P { X = n } = λ k k ! e − λ ( λ > 0 ; k = 0 , 1 , 2 , ⋯ ) \displaystyle P\{X=n\}=\frac{\lambda^k}{k!}e^{-\lambda}\ (\lambda>0;k=0,1,2,\cdots) P{X=n}=k!λke−λ (λ>0;k=0,1,2,⋯)
数字特征:
E ( X ) = D ( X ) = λ E(X)=D(X)=\lambda E(X)=D(X)=λ
proof:
E
(
X
)
=
∑
k
=
0
+
∞
k
⋅
λ
k
k
!
e
−
λ
=
∑
k
=
1
+
∞
λ
k
(
k
−
1
)
!
e
−
λ
=
λ
e
−
λ
∑
k
=
0
+
∞
λ
k
k
!
=
λ
e
−
λ
e
λ
=
λ
E
(
X
2
)
=
∑
k
=
0
+
∞
k
2
⋅
λ
k
k
!
e
−
λ
=
∑
k
=
1
+
∞
k
2
⋅
λ
k
k
!
e
−
λ
=
λ
e
−
λ
∑
k
=
1
+
∞
k
⋅
λ
k
−
1
(
k
−
1
)
!
=
λ
e
−
λ
[
∑
k
=
1
+
∞
λ
k
(
k
−
1
)
!
]
′
=
λ
e
−
λ
[
λ
∑
k
=
0
+
∞
λ
k
k
!
]
′
=
λ
e
−
λ
(
λ
e
λ
)
′
=
λ
(
1
+
λ
)
D
(
X
)
=
E
(
X
2
)
−
[
E
(
X
)
]
2
=
λ
(
1
+
λ
)
−
λ
2
=
λ
\begin{aligned} E(X)&=\sum_{k=0}^{+\infty}k\cdot \frac{\lambda^k}{k!}e^{-\lambda} =\sum_{k=1}^{+\infty}\frac{\lambda^k}{(k-1)!}e^{-\lambda} \\&=\lambda e^{-\lambda}\sum_{k=0}^{+\infty}\frac{\lambda^k}{k!} =\lambda e^{-\lambda}e^{\lambda}=\lambda \\E(X^2)&=\sum_{k=0}^{+\infty}k^2\cdot\frac{\lambda^k}{k!}e^{-\lambda} =\sum_{k=1}^{+\infty}k^2\cdot\frac{\lambda^k}{k!}e^{-\lambda} =\lambda e^{-\lambda}\sum_{k=1}^{+\infty}k\cdot\frac{\lambda^{k-1}}{(k-1)!} \\&=\lambda e^{-\lambda}\left[\sum_{k=1}^{+\infty}\frac{\lambda^{k}}{(k-1)!}\right]' =\lambda e^{-\lambda}\left[\lambda\sum_{k=0}^{+\infty}\frac{\lambda^{k}}{k!}\right]' \\&=\lambda e^{-\lambda}(\lambda e^{\lambda})'=\lambda(1+\lambda) \\D(X)&=E(X^2)-[E(X)]^2=\lambda(1+\lambda)-\lambda^2=\lambda \end{aligned}
E(X)E(X2)D(X)=k=0∑+∞k⋅k!λke−λ=k=1∑+∞(k−1)!λke−λ=λe−λk=0∑+∞k!λk=λe−λeλ=λ=k=0∑+∞k2⋅k!λke−λ=k=1∑+∞k2⋅k!λke−λ=λe−λk=1∑+∞k⋅(k−1)!λk−1=λe−λ[k=1∑+∞(k−1)!λk]′=λe−λ[λk=0∑+∞k!λk]′=λe−λ(λeλ)′=λ(1+λ)=E(X2)−[E(X)]2=λ(1+λ)−λ2=λ
泊松积分:
∫ 0 + ∞ e − x 2 d x = π 2 \displaystyle \int_{0}^{+\infty}e^{-x^2}\mathrm{d}x=\frac{\sqrt{\pi}}{2} ∫0+∞e−x2dx=2π
proof1:(标准正态分布)
∫
0
+
∞
1
2
π
e
−
1
2
x
2
d
x
=
t
=
x
2
∫
0
+
∞
1
π
e
−
t
2
d
t
=
1
2
⇒
∫
0
+
∞
e
−
x
2
d
x
=
π
2
\int_0^{+\infty}\frac{1}{\sqrt{2\pi}}e^{-\frac{1}{2}x^2}\mathrm{d}x\xlongequal{t=\frac{x}{\sqrt{2}}}\int_0^{+\infty}\frac{1}{\sqrt{\pi}}e^{-t^2}\mathrm{d}t=\frac{1}{2} \ \Rightarrow \int_{0}^{+\infty}e^{-x^2}\mathrm{d}x=\frac{\sqrt{\pi}}{2}
∫0+∞2π1e−21x2dxt=2x∫0+∞π1e−t2dt=21 ⇒∫0+∞e−x2dx=2π
proof2:(Gamma函数与余元公式)
∫
0
+
∞
e
−
x
2
d
x
=
1
2
Γ
(
1
2
)
=
1
2
Γ
(
1
2
)
2
=
1
2
Γ
(
1
2
)
Γ
(
1
−
1
2
)
=
1
2
π
sin
(
π
2
)
=
π
2
\begin{aligned} \int_0^{+\infty}e^{-x^2}\mathrm{d}x &=\frac{1}{2}\Gamma\left(\frac{1}{2}\right)=\frac{1}{2}\sqrt{\Gamma\left(\frac{1}{2}\right)^2}=\frac{1}{2}\sqrt{\Gamma\left(\frac{1}{2}\right)\Gamma\left(1-\frac{1}{2}\right)} \\&=\frac{1}{2}\sqrt{\frac{\pi}{\sin\left(\frac{\pi}{2}\right)}}=\frac{\sqrt{\pi}}{2} \end{aligned}
∫0+∞e−x2dx=21Γ(21)=21Γ(21)2=21Γ(21)Γ(1−21)=21sin(2π)π=2π
proof3:(累次积分)
∫
0
+
∞
e
−
x
2
d
x
=
∫
0
+
∞
e
−
y
2
d
y
=
∫
0
+
∞
e
−
x
2
d
x
∫
0
+
∞
e
−
y
2
d
y
=
∬
[
0
,
+
∞
]
2
e
−
(
x
2
+
y
2
)
d
x
d
y
=
∫
0
π
2
d
θ
∫
0
+
∞
e
−
r
2
r
d
r
=
t
=
r
2
1
2
∫
0
π
2
d
θ
∫
0
+
∞
e
−
t
d
t
=
1
2
∫
0
π
2
d
θ
∫
0
+
∞
e
−
t
d
t
=
1
2
∫
0
π
2
d
θ
=
π
2
\begin{aligned} \int_0^{+\infty}e^{-x^2}\mathrm{d}x&=\int_0^{+\infty}e^{-y^2}\mathrm{d}y =\sqrt{\int_0^{+\infty}e^{-x^2}\mathrm{d}x\int_0^{+\infty}e^{-y^2}\mathrm{d}y} \\&=\sqrt{\iint_{[0,+\infty]^2}e^{-(x^2+y^2)}\mathrm{d}x\mathrm{d}y} =\sqrt{\int_0^{\frac{\pi}{2}}\mathrm{d}\theta\int_0^{+\infty}e^{-r^2}r\mathrm{d}r} \\&\xlongequal{t=r^2}\sqrt{\frac{1}{2}\int_0^{\frac{\pi}{2}}\mathrm{d}\theta\int_0^{+\infty}e^{-t}\mathrm{d}t} =\sqrt{\frac{1}{2}\int_0^{\frac{\pi}{2}}\mathrm{d}\theta\int_0^{+\infty}e^{-t}\mathrm{d}t} \\&=\sqrt{\frac{1}{2}\int_0^{\frac{\pi}{2}}\mathrm{d}\theta}=\frac{\sqrt{\pi}}{2} \end{aligned}
∫0+∞e−x2dx=∫0+∞e−y2dy=∫0+∞e−x2dx∫0+∞e−y2dy=∬[0,+∞]2e−(x2+y2)dxdy=∫02πdθ∫0+∞e−r2rdrt=r221∫02πdθ∫0+∞e−tdt=21∫02πdθ∫0+∞e−tdt=21∫02πdθ=2π
独立可加性:
X 1 , 2 ∼ P ( λ 1 , 2 ) 且独立 ⇒ X 1 + X 2 ∼ P ( λ 1 + λ 2 ) X_{1,2}\thicksim P(\lambda_{1,2}) \ \text{且独立}\ \Rightarrow\ X_1+X_2\thicksim P(\lambda_1+\lambda_2) X1,2∼P(λ1,2) 且独立 ⇒ X1+X2∼P(λ1+λ2)
3.1.2 连续型r.v.
3.1.2.1 均匀分布 X ∼ U ( a , b ) X\thicksim U(a,b) X∼U(a,b)
概率密度与分布函数:
f ( x ) = { 1 b − a , a < x < b 0 , o t h e r w i s e ⇒ F ( x ) = { 0 , x < a ∫ a x d t b − a = x − a b − a , a ⩽ x < b 1 , x ⩾ b f(x)=\left\{\begin{array}{l} \displaystyle \frac{1}{b-a}\ ,\ a<x<b\\ \\ 0\qquad\ ,\ otherwise\\ \end{array}\right. \Rightarrow F(x)=\left\{\begin{array}{l} \ 0\qquad\qquad\qquad\qquad\ ,x<a\\ \displaystyle\int_a^x\frac{\mathrm{d}t}{b-a}=\frac{x-a}{b-a}\quad ,a\leqslant x<b\\ \ 1\qquad\qquad\qquad\qquad\ , x\geqslant b\\ \end{array}\right. f(x)=⎩ ⎨ ⎧b−a1 , a<x<b0 , otherwise⇒F(x)=⎩ ⎨ ⎧ 0 ,x<a∫axb−adt=b−ax−a,a⩽x<b 1 ,x⩾b
数字特征:
E ( X ) = a + b 2 , D ( X ) = ( b − a ) 2 12 E(X)=\frac{a+b}{2}\ ,\ D(X)=\frac{(b-a)^2}{12} E(X)=2a+b , D(X)=12(b−a)2
proof:
E
(
X
)
=
∫
a
b
x
⋅
1
b
−
a
d
x
=
a
+
b
2
E
(
X
2
)
=
∫
a
b
x
2
⋅
1
b
−
a
d
x
=
b
2
+
a
b
+
a
2
3
D
(
X
)
=
E
(
X
2
)
−
[
E
(
X
)
]
2
=
(
b
−
a
)
2
12
\begin{aligned} E(X)&=\int_a^b x\cdot\frac{1}{b-a}\mathrm{d}x=\frac{a+b}{2} \\E(X^2)&=\int_a^b x^2\cdot\frac{1}{b-a}\mathrm{d}x=\frac{b^2+ab+a^2}{3} \\D(X)&=E(X^2)-[E(X)]^2=\frac{(b-a)^2}{12} \end{aligned}
E(X)E(X2)D(X)=∫abx⋅b−a1dx=2a+b=∫abx2⋅b−a1dx=3b2+ab+a2=E(X2)−[E(X)]2=12(b−a)2
3.1.2.2 指数分布 X ∼ E ( λ ) X\thicksim E(\lambda) X∼E(λ)
概率密度与分布函数:
f ( x ) = { λ e − λ x , x > 0 0 , x ⩽ 0 ⇒ F ( x ) = { ∫ 0 x λ e − λ t d t = 1 − e − λ x , x > 0 0 , x ⩽ 0 ( λ > 0 ) f(x)=\left\{\begin{array}{l} \lambda e^{-\lambda x},x>0\\ 0\qquad,x\leqslant 0 \end{array}\right. \Rightarrow\ F(x)=\left\{\begin{array}{l} \displaystyle\int_0^x\lambda e^{-\lambda t}\mathrm{d}t=1-e^{-\lambda x},x>0\\ 0\qquad\qquad\qquad\qquad\quad\ ,x\leqslant 0\\ \end{array}\right.(\lambda >0) f(x)={λe−λx,x>00,x⩽0⇒ F(x)=⎩ ⎨ ⎧∫0xλe−λtdt=1−e−λx,x>00 ,x⩽0(λ>0)
数字特征:
E ( X ) = 1 λ , D ( X ) = 1 λ 2 \displaystyle E(X)=\frac{1}{\ \lambda\ }\ ,\ D(X)=\frac{1}{\ \lambda^2\ } E(X)= λ 1 , D(X)= λ2 1
proof:
E
(
X
)
=
∫
0
+
∞
x
⋅
λ
e
−
λ
x
d
x
=
1
λ
∫
0
+
∞
(
λ
x
)
e
−
(
λ
x
)
d
(
λ
x
)
=
t
=
λ
x
−
1
λ
[
t
e
−
t
∣
0
+
∞
−
∫
0
+
∞
e
−
t
d
t
]
=
1
λ
E
(
X
2
)
=
1
λ
2
∫
0
+
∞
(
λ
x
)
2
e
−
(
λ
x
)
d
(
λ
x
)
=
t
=
λ
x
2
λ
2
∫
0
+
∞
t
e
−
t
d
t
=
2
λ
2
D
(
X
)
=
E
(
X
2
)
−
[
E
(
X
)
]
2
=
1
λ
2
\begin{aligned} E(X)&=\int_0^{+\infty}x\cdot\lambda e^{-\lambda x}\mathrm{d}x=\frac{1}{\lambda}\int_0^{+\infty}(\lambda x)e^{-(\lambda x)}\mathrm{d}(\lambda x) \\&\xlongequal{t=\lambda x}-\frac{1}{\lambda}\left[te^{-t}|_0^{+\infty}-\int_0^{+\infty}e^{-t}\mathrm{d}t\right] =\frac{1}{\lambda} \\E(X^2)&=\frac{1}{\lambda^2}\int_0^{+\infty}(\lambda x)^2e^{-(\lambda x)}\mathrm{d}(\lambda x) \xlongequal{t=\lambda x}\frac{2}{\lambda^2}\int_0^{+\infty}te^{-t}\mathrm{d}t=\frac{2}{\lambda^2} \\D(X)&=E(X^2)-[E(X)]^2=\frac{1}{\lambda^2} \end{aligned}
E(X)E(X2)D(X)=∫0+∞x⋅λe−λxdx=λ1∫0+∞(λx)e−(λx)d(λx)t=λx−λ1[te−t∣0+∞−∫0+∞e−tdt]=λ1=λ21∫0+∞(λx)2e−(λx)d(λx)t=λxλ22∫0+∞te−tdt=λ22=E(X2)−[E(X)]2=λ21
3.1.2.3 正态分布(高斯分布)
3.1.2.3.1 一般正态 X ∼ N ( μ , σ 2 ) X\thicksim N(\mu,\sigma^2) X∼N(μ,σ2)
密度与分布:
f ( x ) = 1 2 π σ e − 1 2 ( x − μ σ ) 2 , x ∈ R ⇒ F ( x ) = 1 2 π σ ∫ − ∞ x e − 1 2 ( t − μ σ ) 2 d t f(x)=\frac{1}{\sqrt{2\pi}\sigma}e^{-\frac{1}{2}\left(\frac{x-\mu}{\sigma}\right)^2},x\in\mathbb{R} \ \Rightarrow\ F(x)=\frac{1}{\sqrt{2\pi}\sigma}\int_{-\infty}^{x}e^{-\frac{1}{2}\left(\frac{t-\mu}{\sigma}\right)^2}\mathrm{d}t f(x)=2πσ1e−21(σx−μ)2,x∈R ⇒ F(x)=2πσ1∫−∞xe−21(σt−μ)2dt
数字特征:
E ( X ) = μ , D ( X ) = σ 2 E(X)=\mu\ ,\ D(X)=\sigma^2 E(X)=μ , D(X)=σ2
proof:
E
(
X
)
=
∫
−
∞
+
∞
x
⋅
1
2
π
σ
e
−
1
2
(
x
−
μ
σ
)
2
d
x
=
y
=
x
−
μ
σ
1
2
π
∫
−
∞
+
∞
(
σ
y
+
μ
)
e
−
1
2
y
2
d
y
=
μ
2
π
∫
−
∞
+
∞
e
−
1
2
y
2
d
y
=
μ
π
⋅
2
∫
0
+
∞
(
y
2
)
2
⋅
1
2
−
1
e
−
(
y
2
)
2
d
(
y
2
)
=
μ
π
Γ
(
1
2
)
=
μ
π
⋅
π
=
μ
\begin{aligned} E(X)&=\int_{-\infty}^{+\infty}x\cdot\frac{1}{\sqrt{2\pi}\sigma}e^{-\frac{1}{2}\left(\frac{x-\mu}{\sigma}\right)^2}\mathrm{d}x \xlongequal{y=\frac{x-\mu}{\sigma}}\frac{1}{\sqrt{2\pi}}\int_{-\infty}^{+\infty}(\sigma y+\mu)e^{-\frac{1}{2}y^2}\mathrm{d}y \\&=\frac{\mu}{\sqrt{2\pi}}\int_{-\infty}^{+\infty}e^{-\frac{1}{2}y^2}\mathrm{d}y =\frac{\mu}{\sqrt{\pi}}\cdot 2\int_0^{+\infty}\left(\frac{y}{\sqrt{2}}\right)^{2\cdot \frac{1}{2}-1}e^{-\left(\frac{y}{\sqrt{2}}\right)^2}\mathrm{d}\left(\frac{y}{\sqrt{2}}\right) \\&=\frac{\mu}{\sqrt{\pi}}\Gamma\left(\frac{1}{2}\right)=\frac{\mu}{\sqrt{\pi}}\cdot \sqrt{\pi}=\mu \end{aligned}
E(X)=∫−∞+∞x⋅2πσ1e−21(σx−μ)2dxy=σx−μ2π1∫−∞+∞(σy+μ)e−21y2dy=2πμ∫−∞+∞e−21y2dy=πμ⋅2∫0+∞(2y)2⋅21−1e−(2y)2d(2y)=πμΓ(21)=πμ⋅π=μ
3.1.2.3.2 标准正态 X ∼ N ( 0 , 1 ) X\thicksim N(0,1) X∼N(0,1)
密度与分布:
φ ( x ) = 1 2 π e − 1 2 x 2 , x ∈ R ⇒ F ( x ) = u = t − μ σ Φ ( x ) = 1 2 π ∫ − ∞ x e − 1 2 u 2 d u \displaystyle \varphi(x)=\frac{1}{\sqrt{2\pi}}e^{-\frac{1}{2}x^2},x\in\mathbb{R} \ \Rightarrow\ F(x)\xlongequal{u=\frac{t-\mu}{\sigma}}\Phi(x)=\frac{1}{\sqrt{2\pi}}\int_{-\infty}^x e^{-\frac{1}{2}u^2}\mathrm{d}u φ(x)=2π1e−21x2,x∈R ⇒ F(x)u=σt−μΦ(x)=2π1∫−∞xe−21u2du
3.1.2.4 伽马分布 X ∼ G a ( α , λ ) X\sim Ga(\alpha,\lambda) X∼Ga(α,λ)
3.1.2.5 贝塔分布 X ∼ B e ( a , b ) X\sim Be(a,b) X∼Be(a,b)
3.2 二维r.v.
3.2.1 二维均匀分布 ( X , Y ) ∼ U ( D ) (X,Y)\thicksim U(D) (X,Y)∼U(D)
概率密度函数:
f ( x , y ) = { 1 ∬ D d x d y , ( x , y ) ∈ D 0 , o t h e r w i s e f(x,y)=\left\{\begin{array}{l} \displaystyle \frac{1}{\iint_{_D}\mathrm{d}x\mathrm{d}y}\ ,\ (x,y)\in D\\ \\ 0 \qquad\quad\ \ \ \ ,\ otherwise\\ \end{array}\right. f(x,y)=⎩ ⎨ ⎧∬Ddxdy1 , (x,y)∈D0 , otherwise
3.2.2 二维正态分布 ( X , Y ) ∼ N ( μ 1 , μ 2 ; σ 1 2 , σ 2 2 ; ρ X Y ) (X,Y)\thicksim N(\mu_1,\mu_2;\sigma_1^2,\sigma_2^2;\rho_{_{XY}}) (X,Y)∼N(μ1,μ2;σ12,σ22;ρXY)
概率密度函数:
f ( x , y ) = 1 2 π σ 1 σ 2 1 − ρ 2 exp { 1 2 ( 1 − ρ 2 ) [ ( x − μ 1 ) 2 σ 1 2 − 2 ρ ( x − μ 1 ) ( x − μ 2 ) σ 1 σ 2 + ( x − μ 2 ) 2 σ 2 2 ] } ( − ∞ < x , y < + ∞ ) \displaystyle f(x,y)=\frac{1}{2\pi\sigma_1\sigma_2\sqrt{1-\rho^2}}\exp\left\{\frac{1}{2(1-\rho^2)}\left[\frac{(x-\mu_1)^2}{\sigma_1^2}-2\rho\frac{(x-\mu_1)(x-\mu_2)}{\sigma_1\sigma_2}+\frac{(x-\mu_2)^2}{\sigma_2^2}\right]\right\} \\(-\infty<x,y<+\infty) f(x,y)=2πσ1σ21−ρ21exp{2(1−ρ2)1[σ12(x−μ1)2−2ρσ1σ2(x−μ1)(x−μ2)+σ22(x−μ2)2]}(−∞<x,y<+∞)
性质:
( μ 1 , μ 2 ; σ 1 2 , σ 2 2 ; ρ X Y ) = ( E X , E Y ; D X , D Y ; c o v ( X , Y ) D X D Y ) \displaystyle (\mu_1,\mu_2;\sigma_1^2,\sigma_2^2;\rho_{_{XY}})=\left(EX,EY;DX,DY;\frac{cov(X,Y)}{\sqrt{DX}\sqrt{DY}}\right) (μ1,μ2;σ12,σ22;ρXY)=(EX,EY;DX,DY;DXDYcov(X,Y))
(1) 独立 ⇒ ⇍ \begin{array}{c} \Rightarrow\\ \nLeftarrow\\ \end{array} ⇒⇍不相关;不相关即二者没有线性关系,可能有其他关系
(2) 若 ( X , Y ) ∼ N (X,Y)\thicksim N (X,Y)∼N,则 X , Y X,Y X,Y独立 ⇔ X , Y \Leftrightarrow X,Y ⇔X,Y 不相关 ⇔ \ \Leftrightarrow ⇔ 相关系数 ρ X Y = 0 \rho_{_{XY}}=0 ρXY=0(1) ( X , Y ) ∼ N ( μ 1 , μ 2 ; σ 1 2 , σ 2 2 ; ρ X Y ) ⇒ ⇍ { X ∼ N ( μ 1 , σ 1 2 ) Y ∼ N ( μ 2 , σ 2 2 ) (X,Y)\thicksim N(\mu_1,\mu_2;\sigma_1^2,\sigma_2^2;\rho_{_{XY}}) \begin{array}{c}\Rightarrow\\ \nLeftarrow\\ \end{array} \left\{\begin{array}{l} X\thicksim N(\mu_1,\sigma_1^2)\\ Y\thicksim N(\mu_2,\sigma_2^2)\\ \end{array}\right. (X,Y)∼N(μ1,μ2;σ12,σ22;ρXY)⇒⇍{X∼N(μ1,σ12)Y∼N(μ2,σ22)
(2) { X ∼ N ( μ 1 , σ 1 2 ) Y ∼ N ( μ 2 , σ 2 2 ) \left\{\begin{array}{l} X\thicksim N(\mu_1,\sigma_1^2)\\ Y\thicksim N(\mu_2,\sigma_2^2)\\ \end{array}\right. {X∼N(μ1,σ12)Y∼N(μ2,σ22)且独立 ⇒ ( X , Y ) ∼ N ( μ 1 , μ 2 ; σ 1 2 , σ 2 2 ; 0 ) \Rightarrow (X,Y)\sim N(\mu_1,\mu_2;\sigma_1^2,\sigma_2^2;0) ⇒(X,Y)∼N(μ1,μ2;σ12,σ22;0)两个正态分布的线性组合仍然服从正态分布,即
若 { X ∼ N ( μ 1 , σ 1 2 ) Y ∼ N ( μ 2 , σ 2 2 ) \left\{\begin{array}{l} X\thicksim N(\mu_1,\sigma_1^2)\\ Y\thicksim N(\mu_2,\sigma_2^2)\\ \end{array}\right. {X∼N(μ1,σ12)Y∼N(μ2,σ22),则 ∀ a , b \forall a,b ∀a,b 不全为零, a X + b Y ∼ N ( a μ 1 + b μ 2 , a 2 σ 1 2 + b 2 σ 2 2 + 2 a b ρ X Y σ 1 σ 2 ) aX+bY\thicksim N(a\mu_1+b\mu_2,a^2\sigma_1^2+b^2\sigma_2^2+2ab\rho_{_{XY}}\sigma_1\sigma_2) aX+bY∼N(aμ1+bμ2,a2σ12+b2σ22+2abρXYσ1σ2)
3.3 抽样分布
3.3.1 χ \chi χ 分布
定义:
{ X i ( i = 1 , ⋯ , n ) 独立 X i ∼ N ( 0 , 1 ) ⇒ χ 2 = ∑ k = 1 n X i 2 ∼ χ 2 ( n ) \left\{\begin{array}{l} X_i(i=1,\cdots,n)\ 独立\\ X_i\thicksim N(0,1)\\ \end{array}\right. \Rightarrow \chi^2=\sum\limits_{k=1}^{n}X_i^2\thicksim \chi^2(n) {Xi(i=1,⋯,n) 独立Xi∼N(0,1)⇒χ2=k=1∑nXi2∼χ2(n)
独立可加性:
χ 1 2 ∼ χ 2 ( n 1 ) , χ 2 2 ∼ χ 2 ( n 2 ) \chi_1^2\thicksim \chi^2(n_1),\chi_2^2\thicksim \chi^2(n_2) χ12∼χ2(n1),χ22∼χ2(n2) 且独立 ⇒ χ 1 2 + χ 1 2 ∼ χ 2 ( n 1 + n 2 ) \Rightarrow \chi_1^2+\chi_1^2\thicksim \chi^2(n_1+n_2) ⇒χ12+χ12∼χ2(n1+n2)
数字特征:
E ( χ 2 ( n ) ) = n , D ( χ 2 ( n ) ) = 2 n E(\chi^2(n))=n\ ,\ D(\chi^2(n))=2n E(χ2(n))=n , D(χ2(n))=2n
推论:
X ∼ N ( 0 , 1 ) ⇒ X 2 ∼ χ 2 ( 1 ) X\thicksim N(0,1)\Rightarrow X^2\sim \chi^2(1) X∼N(0,1)⇒X2∼χ2(1)
自由度为2的卡方分布为参数0.5的指数分布: χ 2 ( 2 ) = E ( 1 2 ) \displaystyle \chi^2(2)=E\left(\frac{1}{2}\right) χ2(2)=E(21)
proof corollary2:
设
X
,
Y
∼
N
(
0
,
1
)
X,Y\thicksim N(0,1)
X,Y∼N(0,1) 且相互独立,
Z
=
X
2
+
Y
2
∼
χ
2
(
2
)
Z=X^2+Y^2\thicksim \chi^2(2)
Z=X2+Y2∼χ2(2),则
F Z ( z ) = P { Z ⩽ z } = P { X 2 + Y 2 ⩽ z } = { ∬ x 2 + y 2 ⩽ z 1 2 π e − x 2 + y 2 2 d x d y = 1 − e − z 2 , z > 0 0 , z ⩽ 0 \begin{aligned} F_{_Z}(z)&=P\{Z\leqslant z\}=P\{X^2+Y^2\leqslant z\} \\&=\begin{cases} \displaystyle\iint\limits_{x^2+y^2\leqslant z}\frac{1}{2\pi}e^{-\frac{x^2+y^2}{2}}\mathrm{d}x\mathrm{d}y=1-e^{-\frac{z}{2}}& ,z>0\\ \\ \qquad 0& ,z\leqslant 0\\ \end{cases} \end{aligned} FZ(z)=P{Z⩽z}=P{X2+Y2⩽z}=⎩ ⎨ ⎧x2+y2⩽z∬2π1e−2x2+y2dxdy=1−e−2z0,z>0,z⩽0
故
Z ∼ E ( 1 2 ) = χ 2 ( 2 ) Z\thicksim E\left(\frac{1}{2}\right)=\chi^2(2) Z∼E(21)=χ2(2)
分位点:
P { χ 2 > χ α 2 ( n ) } = ∫ χ α 2 ( n ) + ∞ f ( t ) d t = α = P { χ 2 ⩽ χ 1 − α 2 ( n ) } \displaystyle P\{\chi^2>\chi_{\alpha}^2(n)\}=\int_{\chi_{\alpha}^2(n)}^{+\infty}f(t)\mathrm{d}t=\alpha=P\{\chi^2\leqslant\chi_{1-\alpha}^2(n)\} P{χ2>χα2(n)}=∫χα2(n)+∞f(t)dt=α=P{χ2⩽χ1−α2(n)}
3.3.2 t t t 分布
定义:
{ X ∼ N ( 0 , 1 ) Y ∼ χ 2 ( n ) X , Y 独立 ⇒ t = X Y / n ∼ t ( n ) \left\{\begin{array}{l} X\thicksim N(0,1)\\ Y\thicksim \chi^2(n)\\ X,Y\ 独立\\ \end{array}\right. \Rightarrow\ t=\frac{X}{\sqrt{Y/n}}\thicksim t(n) ⎩ ⎨ ⎧X∼N(0,1)Y∼χ2(n)X,Y 独立⇒ t=Y/nX∼t(n)
性质:
E ( t ( n ) ) = 0 E(t(n))=0 E(t(n))=0
密度函数为偶函数, n n n 足够大时近似分布于 N ( 0 , 1 ) N(0,1) N(0,1).
分位点:
P { t > t α ( n ) } = ∫ t α ( n ) + ∞ f ( t ) d t = α = P { t ⩽ t 1 − α ( n ) = − t α ( n ) } \displaystyle P\{t>t_{\alpha}(n)\}=\int_{t_{\alpha}(n)}^{+\infty}f(t)\mathrm{d}t=\alpha=P\{t\leqslant t_{1-\alpha}(n)=-t_{\alpha}(n)\} P{t>tα(n)}=∫tα(n)+∞f(t)dt=α=P{t⩽t1−α(n)=−tα(n)}
3.3.3 F F F 分布
定义:
{ X ∼ χ 2 ( n 1 ) Y ∼ χ 2 ( n 2 ) X , Y 独立 ⇒ F = X / n 1 Y / n 2 ∼ F ( n 1 , n 2 ) \left\{\begin{array}{l} X\thicksim \chi^2(n_1)\\ Y\thicksim \chi^2(n_2)\\ X,Y\ 独立\\ \end{array}\right. \Rightarrow\ F=\frac{X/n_1}{Y/n_2}\thicksim F(n_1,n_2) ⎩ ⎨ ⎧X∼χ2(n1)Y∼χ2(n2)X,Y 独立⇒ F=Y/n2X/n1∼F(n1,n2)
性质:
F ∼ F ( n 1 , n 2 ) ⇒ 1 F ∼ F ( n 1 , n 2 ) \displaystyle F\thicksim F(n_1,n_2)\Rightarrow \frac{1}{F}\thicksim F(n_1,n_2) F∼F(n1,n2)⇒F1∼F(n1,n2)
t = X Y / n ∼ t ( n ) ⇒ t 2 = X / 1 Y / n ∼ F ( 1 , n ) \displaystyle t=\frac{X}{\sqrt{Y/n}}\thicksim t(n)\Rightarrow t^2=\frac{X/1}{Y/n}\thicksim F(1,n) t=Y/nX∼t(n)⇒t2=Y/nX/1∼F(1,n)
分位点:
P { F > F α ( n 1 , n 2 ) } = ∫ F α ( n 1 , n 2 ) + ∞ f ( t ) d t = α \displaystyle P\{F>F_{\alpha}(n_1,n_2)\}=\int_{F_{\alpha}(n_1,n_2)}^{+\infty}f(t)\mathrm{d}t=\alpha P{F>Fα(n1,n2)}=∫Fα(n1,n2)+∞f(t)dt=α
F α ( n 1 , n 2 ) = 1 F 1 − α ( n 1 , n 2 ) \displaystyle F_{\alpha}(n_1,n_2)=\frac{1}{F_{1-\alpha}(n_1,n_2)} Fα(n1,n2)=F1−α(n1,n2)1