I am new user in R. Please help me!
I have 1114 observations with 8 assets. For example, I have fitted a multivariate DCC-GARCH model to the first 1000 data points and I want to do 1-ahead forecast for 3 periods later such as
1) Data[1:1000,] In-sample data, forecast for Data[1001,]
2) Data[1:1001,] In-sample data, forecast for Data[1002,]
3) Data[1:1002,] In-sample data, forecast for Data[1003,]
Below is my reproducible code:-
# load libraries
library(rugarch)
library(rmgarch)
library(FinTS)
library(tseries)
library (fPortfolio)
data(dji30retw)
for (i in 1:3) {
Dat.Initial = dji30retw[, 1:8, drop = FALSE]
Dat <- Dat.Initial[1:(1000+(i-1)), ]
#Fitting the data
uspec = ugarchspec(mean.model = list(armaOrder = c(0,0)), variance.model = list(garchOrder = c(1,1), model = "sGARCH"), distribution.model = "norm")
spec1 = dccspec(uspec = multispec( replicate(8, uspec)), dccOrder = c(1,1), distribution = "mvnorm")
fit1 <- list()
fit1[[i]] = dccfit(spec1, data = Dat, out.sample = 1, fit.control = list(eval.se=T))
#Out of sample forecasting
dcc.focast <- list()
dcc.focast[[i]]=dccforecast(fit1[[i]], n.ahead = 1, n.roll = 0)
print(dcc.focast[[i]])
}
The codes works perfectly. I can now get my dcc.focast values. But why is it, if I execute
dcc.focast[[1]]
NULL
it gives me "NULL". Shouldn't it produce the same answer as the "print(dcc.focast[[i]])" as in the loop?
The problem here, it only gives me dcc.focast[[3]]. The rest are are "NULL". What is the mistake that I have done? Anyone can help to explain?
Modify your code to define your lists outside the for. You are deleting the previous computations when redefining your lists inside the loop.
data(dji30retw)
fit1 <- list()
dcc.focast <- list()
for (i in 1:3) {#i=1
Dat.Initial = dji30retw[, 1:8, drop = FALSE]
Dat <- Dat.Initial[1:(1000+(i-1)), ]
#Fitting the data
uspec = ugarchspec(mean.model = list(armaOrder = c(0,0)), variance.model = list(garchOrder = c(1,1), model = "sGARCH"), distribution.model = "norm")
spec1 = dccspec(uspec = multispec( replicate(8, uspec)), dccOrder = c(1,1), distribution = "mvnorm")
fit1[[i]] = dccfit(spec1, data = Dat, out.sample = 1, fit.control = list(eval.se=T))
#Out of sample forecasting
dcc.focast[[i]]=dccforecast(fit1[[i]], n.ahead = 1, n.roll = 0)
print(dcc.focast[[i]])
}
summary(dcc.focast)
dcc.focast[[1]]
> summary(dcc.focast)
Length Class Mode
[1,] 1 DCCforecast S4
[2,] 1 DCCforecast S4
[3,] 1 DCCforecast S4
> dcc.focast[[1]]
*---------------------------------*
* DCC GARCH Forecast *
*---------------------------------*
Distribution : mvnorm
Model : DCC(1,1)
Horizon : 1
Roll Steps : 0
-----------------------------------
0-roll forecast:
, , 1
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
[1,] 1.0000 0.3378 0.3244 0.2383 0.2730 0.5086 0.3470 0.5073
[2,] 0.3378 1.0000 0.3498 0.5423 0.5910 0.3321 0.2802 0.3658
[3,] 0.3244 0.3498 1.0000 0.2372 0.2771 0.3479 0.2433 0.3625
[4,] 0.2383 0.5423 0.2372 1.0000 0.5746 0.2877 0.2303 0.3593
[5,] 0.2730 0.5910 0.2771 0.5746 1.0000 0.3063 0.2169 0.3101
[6,] 0.5086 0.3321 0.3479 0.2877 0.3063 1.0000 0.3637 0.4307
[7,] 0.3470 0.2802 0.2433 0.2303 0.2169 0.3637 1.0000 0.3976
[8,] 0.5073 0.3658 0.3625 0.3593 0.3101 0.4307 0.3976 1.0000