Hello i'm trying to estimate a correlation matrix. The algorithm seems to work, but i have two questions. How do i test it ( do i have to necessarily perform simulations ?) . And i couldn't anderstand what the argument ts in cdcc_estimation is for.
library(rugarch)
library(xdcclarge)
#load data
mydata= read.csv("returns.csv", header=TRUE)
n<-6
mydata=mydata[,c(21,23,55,58,64,66)]
# Step 1:GARCH Parameter Estimation with rugarch
spec = ugarchspec()
mspec = multispec( replicate(spec, n = n) )
fitlist = multifit(multispec= mspec, data = mydata)
ht<-sigma(fitlist)^2 # les variances conditionnelles
residuals<-residuals(fitlist)
# Step 2:cDCC-GARCH Parameter Estimation with xdcclarge
cDCC<-cdcc_estimation(ini.para=c(0.05,0.93) ,ht ,residuals,method="NLS", ts=1)
#Time varying correlation matrix Rt at time t
(Rt<-matrix(cDCC$cdcc_Rt,n,n))