内容发布更新时间 : 2024/11/19 19:36:05星期一 下面是文章的全部内容请认真阅读。

Copula 汇总表 Copula Gaussian t Gumbel Clayton Frank AMH FGM Joe A12 A14 C(u, v; θ) ?2??(u1),?(u2);??① ?1?1c(u, v; θ) ?2???1(u1),??1(u2);???(u1)?(u2)t2?T??1(u1),T??1(u2);?,??t?(u1)t?(u2)τ ρs 6?arcsin ?26?arcsin ?2θ (-1,1) (-1,1) [1,+?) (0,+?) (-?,+?)\\{0} [-1,1) [-1,1] [1,+?) [1,+?) [1,+?) 2arcsin? ?2arcsin? ?11? ?T2?T(u1),T(u2);?,??② ?1??1?③ exp??w1/?? w?(?lnu1)??(?lnu2)? w?1/? w?u1???u2???1 C(u1,u2;?)?w1/????1? u1u2w2?1/?(lnu1lnu2)1?? 1?12?D1(??)?D2(??)?[1][2] ?(1??)(u1u2)???1w?1/??2 2 ??21?4?D1(??)?1?②[1] ?1?ln?1??e???1??1w1w2? ?wi??e??ui?1? u1u2(1?w)?1 w??(1?u1)(1?u2) u1u2??u1u2(1?u1)(1?u2) ??(e???1)e??(u1?u2)?(e???1)?w1w2?(1?w)32 ?2?u1u2?(1??)w???1?1??(1?2u1)(1?2u2) 2??2(??1)2ln(1??) [3][4]1?23?2?/9 ?/3 1?(w1?w2?w1w2)1/? wi?(1?ui)? 2?w?v2?w2?v?3/2 1?1?2 3? ?1?(w1?w2)1/??wi?(u?1)?1i??1 ???1?(w1?w2)1/?? wi?(ui?1/??1)? 2 1?2?Plackett (??1)?1(w?w2?v)/2 w?1?(??1)(u1?u2) v?4u1u2?(??1) （0,+?)\\{1}

注：①?2??(u1),?(u2);???????1?1??1(u1)???1(u2)??2?s12?2?s1s2?s2?kexp??ds1ds2；②Dk(x)?k?22(1??)x2?1??2??1?x0kxtk，，k = 1,2。 D(?x)?D(x)?dtkkk?1et?1

[1]

Understanding relationships using Copulas pp.10

[2] Stochastic Frontier Models with Correlated Error Components pp.15 [3] Bayesian copula selection pp.814

[4] On the relationship between Spearman’s rho and Kendall’s tau for pairs of continuous random variables

pp.2149

令H为具有边缘分布F、G的联合分布函数，那么存在一个Copula函数C，使得

H(x,y)?C?F(x),G(y)?

如果F，G是连续的，则函数C是唯一的。

(1)

h(x,y)?c?F(x),G(y)?f(x)g(y)

对数正态分布

Copula类型：Gaussian Copula 、Gumbel Copula、Clayton Copula、Frank Copula

(2)

Gaussian Copula

CGa(u1,u2;?)????1?u1??????1?u2???2?x12?2?x1x2?x2?exp???dx1dx2222?1???? 2?1???1??2???1(u1),??1(u2);????2???1(u)?2?2???1(u)??1(u)??2???1(u)?2?1212cGa(u1,u2;?)?exp???22??2(1??)1????

?2???1(u1),??1(u2);???????1(u1)?????1(u2)?1(Continuous Bivariate Distributions pp36)

????1(u)?2?2???1(u)??1(u)????1(u)?21212exp??2?2(1??)?2C(u1,u2;?)?2C(u1,u2;?)?x1?x2?2(x1,x2;?)?c(u1,u2;?)??????u1?u2?x1?x2?u1?u2?(x1)?(x2)2?????1(u1)?????1(u2)?1??2????1(u)?2????1(u)?2???1(u)?2?2???1(u)??1(u)????1(u)?2121212exp???22(1??2)??1??2??????????????1(u)?2?2???1(u)??1(u)????1(u)?2?1212?exp??2??2(1??)??????1(u)?2?1????1(u)?212122?1??exp?????2?exp???2??2??2????????（求

??2???1(u)?2?2???1(u1)??1(u2)??2???1(u)?212exp???2(1??2)1??2?1导）

FGa(y1,y2;?)????1?F1(y1)??????1?F2(y2)???2?x12?2?x1x2?x2??exp??dx1dx2222??1?? 2?1????1??2???1?F1(y1)?,??1?F2(y2)?;???2FGa(y1,y2;?)fGa(y1,y2;?)??cGa?F1(x1),F2(x2)?f1(x1)f2(x2)?y1?y2???1?F1(y1)????1?F2(y2)???2???F1(y1)?,??F2(y2)?;???y1?y1?1?1

??2???1?F1(y1)?,??1?F2(y2)?;??x1???1?F1(y1)?x2???1f1(y1)f2(y2)????1?F1(y1)??????1?F2(y2)??其中

?F2(y2)?，则

fGa(y1,y2;?)??2?x1,x2;??????f1(y1)f2(y2)

?(x1)?(x2)?L(y1,y2)??f1(y1)f(y2)?y1??1??y2??2??dy1dy2x,x;???????212????????(x)?(x)?1??2?12 (3)

?????????????y1??1????1??y2??2????2???2?x1,x2;??dx1dx2????L(u1,u2)?12????????u1?0.5??u2?0.5??2(x1,x2;?)dx1dx2?12??1?100u1u2dCGs(u1,u2;?)?0.5?E(u1)?E(u2)??0.25?

?12?1?100u1u2dCGs(u1,u2;?)?3??S(u1,u2)?(u)?4?111,u20?0CGa(u1,u2;?)dCGa(u1,u2;?)?1

?(x??????L1,x2)?2sin??6?S(u1,u2)???sin??2?(u1,u2)??

Gumbel Copula

CGum(u1,u2;?)?exp???(?lnu1/?1)?(?lnu2)1/????

FGum(y1,y2;?)?exp?????ln?F1(y1/?1/?1)?????ln?F2(y2)?????

cuCGum(u1,u2;?)(lnu1?lnu2)1/??1?1/?1/??1?Gum(1,u2;?)?u1/?1/?2???(?lnu)???1? 1u2?(?lnu1)?(?lnu2)???1(?lnu2)???（Copula 及其在金融分析上的应用，pp22）

（UNDERSTANDING RELATIONSHIPS USING COPULAS）

f?)??2FGum(y1,y2;?)Gum(y1,y2;?y?cGum?F1(y1),F2(y2)?f1(y1)f2(y2)

1?y2??????y1??1??L(y1,y2)????????????y2??2???fGum(y1,y2;?)dy1dy2 2??2????L(x1,x2)?????????x1x2fGum(x1,x2;?)dx1dx2

???????????x1x2cGum??(x1),?(x2);???(x1)?(x2)dx1dx2(4)

11?S(u1,u2)??L(u1,u2)?12?0?0u1u2dCGum(u1,u2;?)?3

?12?1?100CGum(u1,u2;?)du1du2?3Clayton Copula

C,u????Cl(u12;?)?(u1?u2?1)?1/?

FCl(y1,y2;?)???F1(y1)?????F?2(y2)????1??1/

c?2CCl(u1,u2;?)1??Cl(u1,u2;?)??u?(1??)(u1u2)???(u??1?u2?1)?2?1/??u

12（Copula 及其在金融分析上的应用，pp23）

（UNDERSTANDING RELATIONSHIPS USING COPULAS） （Bivariate Distribution Modeling of Tree Diameters and Heights:

Dependency Modeling Using Copulas pp6）

fCl(y1,y2;?)?cCl?F1(y1),F2(y2)?f1(y1)f2(y2)

???????y1??1??y2??2?L(y1,y2)?????????2?????2??fCl(y1,y2;?)dy1dy2 ???L(x1,x2)?????????x1x2fCl(x1,x2;?)dx1dx2

???????????x1x2cCl??(x1),?(x2);???(x1)?(x2)dx1dx2?11S(u1,u2)??L(u1,u2)?12?0?0u1u2dCCl(u1,u2;?)?3

?12?1?100CCl(u1,u2;?)du1du2?3Frank Copula

C1??exp(??u1)?1??exp(??u2)?1??Fr(u1,u2;?)???ln?1?1? ??(??,??)\\{0}?exp(??)??FFr(y1,y2;?)?CFr?F1(y1),F2(y2);??

;?)??2CFr(u1,u2;?)??(e???1)e??(u1?u2)cFr(u1,u2?u? 1?u2?(e???1)?(e??u1?1)(e??u2?1)?2（Copula 及其在金融分析上的应用，pp.24）

(5)