R 中的混合 copula

Question

我想用混合copula進行可靠性分析，現在，在朋友的幫助下，我已經完成了'RVMs_fitted'。現在我想進行概率積分變換（PIT），但是RVINEPIT的function不能使用，因為 RVINEPIT(data，RVM)，這個 RVM 不是 RVINEMATRIX 這是我的代碼：

library(vineclust)
data1 <- read.csv("D:/ASTUDY/Rlanguage/Mix copula/data.csv", header = FALSE)
fit <- vcmm(data = data1, total_comp=3,is_cvine = 0)
print(fit)
summary(fit) 
RVMs_fitted <- list()
RVMs_fitted[[1]] <- VineCopula::RVineMatrix(Matrix=fit$output$vine_structure[,,1],
                                            family=fit$output$bicop_familyset[,,1],
                                            par=fit$output$bicop_param[,,1],
                                            par2=fit$output$bicop_param2[,,1])
RVMs_fitted[[2]] <- VineCopula::RVineMatrix(Matrix=fit$output$vine_structure[,,2],
                                            family=fit$output$bicop_familyset[,,2],
                                            par=fit$output$bicop_param[,,2],
                                            par2=fit$output$bicop_param2[,,2])
RVMs_fitted[[3]] <- VineCopula::RVineMatrix(Matrix=fit$output$vine_structure[,,3],
                                            family=fit$output$bicop_familyset[,,3],
                                            par=fit$output$bicop_param[,,3],
                                            par2=fit$output$bicop_param2[,,3])
RVM<-RVMs_fitted

meanx <- c(0.47,0.508,0.45,0.52,0.48)
sigmax <- c(0.318,0.322,0.296,0.29,0.279)
ux1<-pnorm(x[1],meanx[1],sigmax[1])
ux2<-pnorm(x[2],meanx[2],sigmax[2])
ux3<-pnorm(x[3],meanx[3],sigmax[3])
ux4<-pnorm(x[4],meanx[4],sigmax[4])
ux5<-pnorm(x[5],meanx[5],sigmax[5])
data <- c(ux1,ux2,ux3,ux4,ux5)
du=RVinePIT(data, RVM)
y=t(qnorm(t(du)))


Error: 
 In RVinePIT: RVM has to be an RVineMatrix object.

Answer 1

您在這里有多個問題：

1. RVM 是一個列表。 但是，您嘗試將RVinePIT列表中，而它一次僅適用於一個數據。
2. y也是如此。

我沒有你的數據，但用其他數據試試。

這是代碼（它應該可以工作）：

  library(vineclust)
  library(VineCopula)
data1 <- read.csv("D:/ASTUDY/Rlanguage/Mix copula/data.csv", header = FALSE)
fit <- vcmm(data = data, total_comp=3,is_cvine = 0)
print(fit)
summary(fit) 
RVMs_fitted <- list()
RVMs_fitted[[1]] <- RVineMatrix(Matrix=fit$output$vine_structure[,,1],
                                            family=fit$output$bicop_familyset[,,1],
                                            par=fit$output$bicop_param[,,1],
                                            par2=fit$output$bicop_param2[,,1])
RVMs_fitted[[2]] <- RVineMatrix(Matrix=fit$output$vine_structure[,,2],
                                            family=fit$output$bicop_familyset[,,2],
                                            par=fit$output$bicop_param[,,2],
                                            par2=fit$output$bicop_param2[,,2])
RVMs_fitted[[3]] <- RVineMatrix(Matrix=fit$output$vine_structure[,,3],
                                            family=fit$output$bicop_familyset[,,3],
                                            par=fit$output$bicop_param[,,3],
                                            par2=fit$output$bicop_param2[,,3])
RVM<-RVMs_fitted

meanx <- c(0.47,0.508,0.45,0.52,0.48)
sigmax <- c(0.318,0.322,0.296,0.29,0.279)
ux1<-pnorm(x[1],meanx[1],sigmax[1])
ux2<-pnorm(x[2],meanx[2],sigmax[2])
ux3<-pnorm(x[3],meanx[3],sigmax[3])
ux4<-pnorm(x[4],meanx[4],sigmax[4])
ux5<-pnorm(x[5],meanx[5],sigmax[5])
data <- c(ux1,ux2,ux3,ux4,ux5)### This must be a matrix to work with RVinePIT
du=lapply(1:3, function(i) RVinePIT(data, RVM[[i]]))
y <-lapply(1:3, function(i) t(qnorm(t(du[[i]]))))