u=data(:,1); v=data(:,2); T=length(u); = options optimset('Algorithm','interior-point','Display','iter','TolCon',10^-12,'TolFun',10^-4,'TolX',10^ -6); %优化条件 % 1. Normal Copula kappa1 = corrcoef12(norminv(u),norminv(v)); LL1 = NormalCopula_CL(kappa1,[u,v]); % 2. Clayton's copula lower = 0.0001; theta0 = 1; [ kappa2 LL2] = fmincon('claytonCL',theta0,[],[],[],[],lower,[],[],options,[u,v]); %参数约束非线性 优化方法估计 clayton copula 函数的参数 % 3. Rotated Clayton copula (with tail dep in upper tail instead of lower) lower = 0.0001; theta0 = 1; [ kappa3 LL3] = fmincon('claytonCL',theta0,[],[],[],[],lower,[],[],options,1-[u,v]); % 4. Plackett copula lower = 0.0001; theta0 = 1; [ kappa4 LL4] = fmincon('plackettCL',theta0,[],[],[],[],lower,[],[],options,[u,v]); % LL5 = -3.2721 % 5. Frank copula theta0 = 1; [ kappa5 LL5] = fmincon('frankCL',theta0,[],[],[],[],lower,[],[],options,[u,v]); % 6. Gumbel copula lower = 1.1; theta0 = 2; [ kappa6 LL6] = fmincon('gumbelCL',theta0,[],[],[],[],lower,[],[],options,[u,v]); % 7. Rotated Gumbel copula lower = 1.1; theta0 = 2; [ kappa7 LL7] = fmincon('gumbelCL',theta0,[],[],[],[],lower,[],[],options,1-[u,v]); % 8. Student's t copula 

lower = [-0.999 , 2.1 ]; upper = [ 0.999 , 100 ]; theta0 = [kappa1;10]; [ kappa8 LL8] = fmincon('tcopulaCL',theta0,[],[],[],[],lower,upper,[],options,[u,v]); % 9. Symmetrised Joe-Clayton copula lower = [0 , 0 ]; upper = [ 1 , 1]; theta0 = [0.25;0.25]; [ kappa9 LL9] = fmincon('sym_jc_CL',theta0,[],[],[],[],lower,upper,[],options,[u,v]); LL = [LL1;LL2;LL3;LL4;LL5;LL6;LL7;LL8;LL9]; u,v 为服从 0,1 均匀分布序列，'tcopulaCL'等等为编写好 copula 似然函数(只是提醒不是代码)