TwoCop {TwoCop} | R Documentation |
This function performs the nonparametric test of equality between two copulas proposed by Remillard and Scaillet (2009). The test is based on the Cramer-von-Mises statistic between the two empirical copulas. An approximate p-value is returned.
TwoCop(x, y, Nsim=100, paired=FALSE, alpha=0.95)
x |
|
y |
|
Nsim |
Number of iterations used in the approximation of the p-value. |
paired |
|
alpha |
Level of the calculated VaR. Default is 0.95. |
Details of the method can be found in Remillard and Scaillet (2009).
A list of the following objects:
cvm |
Value of the Cramer-von Mises test statistic. |
pvalue |
pvalue based on the multiplier Monte Carlo method
with |
cvmsim |
Simulated values of the Cramer-von Mises statistic. |
VaR |
|
Bruno Remillard and Jean-Francois Plante
Remillard, B. & Scaillet, O. (2009) Testing for equality between two copulas. Journal of Multivariate Analysis, 100, 377-386.
# Simulating a bivariate normal (copula = independence) X=matrix(rnorm(100),ncol=2) # Simulating a bivriate exponential distribution with a Clayton copula v=runif(50) theta=1 x<-1/(1/runif(50)/v^(theta+1))^(1/(theta+1)) u<-(x^(-theta)-v^(-theta)+1)^(-1/theta) Y=cbind(-log(1-u),-log(1-v)) # Testing equality of the copulas TwoCop(X,Y)$pvalue