离散型
两点分布 B ( 1 , θ ) B(1,\theta) B(1,θ)
f
(
x
;
θ
)
=
P
θ
(
X
=
x
)
=
θ
x
(
1
−
θ
)
1
−
θ
,
x
=
0
,
1
f(x;\theta)=P_\theta(X=x)=\theta^x(1-\theta)^{1-\theta} , x=0,1
f(x;θ)=Pθ(X=x)=θx(1−θ)1−θ,x=0,1
期望:
E
(
X
)
=
θ
E(X)=\theta
E(X)=θ
方差:
D
(
X
)
=
θ
(
1
−
θ
)
D(X)=\theta(1-\theta)
D(X)=θ(1−θ)
特性:当Xi服从B(1, θ \theta θ)那么 ∑ j = 1 n X j 服 从 B ( n , θ ) \sum_{j=1}^{n}X_j服从B(n,\theta) ∑j=1nXj服从B(n,θ)
二项分布 B ( n , θ ) B(n,\theta) B(n,θ)
f
(
x
;
θ
)
=
P
θ
(
X
=
x
)
=
(
k
n
)
θ
x
(
1
−
θ
)
n
−
θ
,
x
=
0
,
1
,
,
,
,
n
;
0
<
θ
<
1
f(x;\theta)=P_\theta(X=x)=(_{k}^{n})\theta^x(1-\theta)^{n-\theta} , x=0,1,,,,n;0<\theta<1
f(x;θ)=Pθ(X=x)=(kn)θx(1−θ)n−θ,x=0,1,,,,n;0<θ<1
期望:
E
(
X
)
=
n
θ
E(X)=n\theta
E(X)=nθ
方差:
D
(
X
)
=
n
θ
(
1
−
θ
)
D(X)=n\theta(1-\theta)
D(X)=nθ(1−θ)
众数:
M
o
d
e
(
X
)
=
(
n
−
1
)
θ
Mode(X)=(n-1)\theta
Mode(X)=(n−1)θ
特性:关于n的可再生性
泊松分布 P ( λ ) P(\lambda) P(λ)
f
(
x
;
λ
)
=
P
λ
(
X
=
x
)
=
e
−
λ
λ
x
x
!
,
x
=
0
,
1
,
,
,
,
;
λ
>
0
f(x;\lambda)=P_\lambda(X=x)=\frac{e^{-\lambda}\lambda^x}{x!} , x=0,1,,,,;\lambda>0
f(x;λ)=Pλ(X=x)=x!e−λλx,x=0,1,,,,;λ>0
期望:
E
(
X
)
=
λ
E(X)=\lambda
E(X)=λ
方差:
D
(
X
)
=
λ
D(X)=\lambda
D(X)=λ
众数:
M
o
d
e
(
X
)
=
λ
Mode(X)=\lambda
Mode(X)=λ
特性:当Xi服从P(\lambda)那么 ∑ j = 1 n X j 服 从 P ( n λ ) \sum_{j=1}^{n}X_j服从P(n\lambda) ∑j=1nXj服从P(nλ)
几何分布 G e ( θ ) Ge(\theta) Ge(θ)
f
(
x
;
θ
)
=
P
θ
(
X
=
x
)
=
θ
(
1
−
θ
)
x
−
1
,
x
=
1
,
2
,
3
,
,
,
;
0
<
θ
<
1
f(x;\theta)=P_\theta(X=x)=\theta(1-\theta)^{x-1} , x=1,2,3,,,;0<\theta<1
f(x;θ)=Pθ(X=x)=θ(1−θ)x−1,x=1,2,3,,,;0<θ<1
期望:
E
(
X
)
=
1
θ
E(X)=\frac {1} {\theta}
E(X)=θ1
方差:
D
(
X
)
=
(
1
−
θ
)
θ
2
D(X)=\frac {(1-\theta)} {\theta^2}
D(X)=θ2(1−θ)
特性:当Xi服从Ge( θ \theta θ)那么 ∑ j = 1 n X j 服 从 N b ( n , θ ) \sum_{j=1}^{n}X_j服从Nb(n,\theta) ∑j=1nXj服从Nb(n,θ)无记忆性
负二项分布 N b ( r , θ ) Nb(r,\theta) Nb(r,θ)
f
(
x
;
θ
)
=
P
θ
(
X
=
x
)
=
(
r
−
1
x
−
1
)
θ
r
(
1
−
θ
)
x
−
r
,
x
=
r
,
r
+
1
,
r
+
2
,
,
,
,
;
0
<
θ
<
1
f(x;\theta)=P_\theta(X=x)=(_{r-1}^{x-1})\theta^r(1-\theta)^{x-r} , x=r,r+1,r+2,,,,;0<\theta<1
f(x;θ)=Pθ(X=x)=(r−1x−1)θr(1−θ)x−r,x=r,r+1,r+2,,,,;0<θ<1
期望:
E
(
X
)
=
r
θ
E(X)=\frac {r} {\theta}
E(X)=θr
方差:
D
(
X
)
=
r
(
1
−
θ
)
θ
2
D(X)=\frac {r(1-\theta)} {\theta^2}
D(X)=θ2r(1−θ)
特性:r有再生性
连续型
正态分布 N ( μ , σ 2 ) N(\mu,\sigma^2) N(μ,σ2)
f
(
x
;
θ
)
=
1
2
π
σ
e
x
p
{
−
(
x
−
μ
)
2
2
σ
2
}
f(x;\theta)=\frac{1}{\sqrt{2\pi}\sigma}exp\{-\frac{(x-\mu)^2}{2\sigma^2}\}
f(x;θ)=2πσ1exp{−2σ2(x−μ)2}
期望:
E
(
X
)
=
μ
E(X)=\mu
E(X)=μ
方差:
D
(
X
)
=
σ
2
D(X)=\sigma^2
D(X)=σ2
众数:
M
o
d
e
(
X
)
=
μ
Mode(X)=\mu
Mode(X)=μ
多元正态分布 N p ( μ , σ 2 ) N_p(\mu,\sigma^2) Np(μ,σ2)
f
(
x
;
θ
)
=
(
2
π
)
−
p
2
∣
Σ
∣
−
1
2
e
x
p
{
−
1
2
(
x
−
μ
)
T
Σ
−
1
(
x
−
μ
)
}
f(x;\theta)=(2\pi)^{-\frac{p}{2}}|\varSigma|^{-\frac{1}{2}}exp\{-\frac{1}{2}(x-\mu)^T\varSigma^{-1}(x-\mu)\}
f(x;θ)=(2π)−2p∣Σ∣−21exp{−21(x−μ)TΣ−1(x−μ)}
期望:
E
(
X
)
=
μ
E(X)=\mu
E(X)=μ
方差:
D
(
X
)
=
Σ
D(X)=\varSigma
D(X)=Σ
众数:
M
o
d
e
(
X
)
=
μ
Mode(X)=\mu
Mode(X)=μ
均匀分布U ( θ 1 , θ 2 ) (\theta_1,\theta_2) (θ1,θ2)
f
(
x
;
θ
)
=
1
θ
1
−
θ
2
,
θ
1
<
θ
2
⊆
R
f(x;\theta)=\frac{1}{\theta_1-\theta_2} ,\theta_1<\theta_2\subseteq\R
f(x;θ)=θ1−θ21,θ1<θ2⊆R
期望:
E
(
X
)
=
θ
1
+
θ
2
2
E(X)=\frac{\theta_1+\theta_2}{2}
E(X)=2θ1+θ2
方差:
D
(
X
)
=
(
θ
2
−
θ
1
)
2
12
D(X)=\frac{(\theta_2-\theta_1)^2}{12}
D(X)=12(θ2−θ1)2
指数分布Exp ( λ ) ( \lambda) (λ)
f
(
x
;
λ
)
=
λ
e
−
λ
x
,
λ
>
0
f(x;\lambda)=\lambda e^{-\lambda x},\lambda>0
f(x;λ)=λe−λx,λ>0
期望:
E
(
X
)
=
1
λ
E(X)=\frac{1}{\lambda}
E(X)=λ1
方差:
D
(
X
)
=
1
λ
2
D(X)=\frac{1}{\lambda^2}
D(X)=λ21
众数: M o d e ( X ) = 0 Mode(X)=0 Mode(X)=0
伽马分布
f
(
x
;
r
,
λ
)
=
λ
r
Γ
(
r
)
x
r
−
1
e
−
λ
x
f(x;r,\lambda)=\frac{\lambda^r}{\Gamma(r)}x^{r-1}e^{-\lambda x}
f(x;r,λ)=Γ(r)λrxr−1e−λx
期望:
E
(
X
)
=
r
λ
E(X)=\frac{r}{\lambda}
E(X)=λr
方差:
D
(
X
)
=
r
λ
2
D(X)=\frac{r}{\lambda^2}
D(X)=λ2r
众数: M o d e ( X ) = r − 1 λ Mode(X)=\frac{r-1}{\lambda} Mode(X)=λr−1
逆伽马分布
f
(
x
;
r
,
λ
)
=
λ
r
Γ
(
r
)
x
−
(
r
+
1
)
e
−
λ
x
f(x;r,\lambda)=\frac{\lambda^r}{\Gamma(r)}x^{-(r+1)}e^{-\frac{\lambda}{x}}
f(x;r,λ)=Γ(r)λrx−(r+1)e−xλ
期望:
E
(
X
)
=
λ
r
−
1
E(X)=\frac{\lambda}{r-1}
E(X)=r−1λ
方差:
D
(
X
)
=
λ
2
(
r
−
1
)
2
(
r
−
2
)
D(X)=\frac{\lambda^2}{(r-1)^2(r-2)}
D(X)=(r−1)2(r−2)λ2
众数: M o d e ( X ) = λ r + 1 Mode(X)=\frac{\lambda}{r+1} Mode(X)=r+1λ
贝塔分布 B e ( α , β ) Be(\alpha,\beta) Be(α,β)
f
(
x
;
r
,
λ
)
=
Γ
(
α
+
β
)
Γ
(
α
)
Γ
(
β
)
x
α
−
1
(
1
−
x
)
β
−
1
f(x;r,\lambda)=\frac{\Gamma(\alpha+\beta)}{\Gamma(\alpha)\Gamma(\beta)}x^{\alpha-1}(1-x)^{\beta-1}
f(x;r,λ)=Γ(α)Γ(β)Γ(α+β)xα−1(1−x)β−1
期望:
E
(
X
)
=
α
α
+
β
E(X)=\frac{\alpha}{\alpha+\beta}
E(X)=α+βα
方差:
D
(
X
)
=
α
β
(
α
+
β
)
2
(
α
+
β
+
1
)
D(X)=\frac{\alpha\beta}{(\alpha+\beta)^2(\alpha+\beta+1)}
D(X)=(α+β)2(α+β+1)αβ
众数: M o d e ( X ) = α − 1 α + β − 2 Mode(X)=\frac{\alpha-1}{\alpha+\beta-2} Mode(X)=α+β−2α−1