Search code examples
roptimizationforecasting

Save or Extract variance covariance matrix output from fitted model in R

I have the following reproducible code. 

library(zoo)
library (rugarch)
library(rmgarch)
data("EuStockMarkets")
EuStockLevel <- as.zoo(EuStockMarkets)[,c("DAX","CAC","FTSE")]
EuStockRet <- diff(log(EuStockLevel))

## GARCH-DCC
    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(3, uspec) ),  dccOrder = c(1,1),  distribution = "mvnorm")
    fit1 = dccfit(spec1, data = EuStockRet, fit.control = list(eval.se=T))

#Forecasting 
    dcc.focast=dccforecast(fit1, n.ahead = 1, n.roll = 0)
    print(dcc.focast)

The output will be as follows:- 

*---------------------------------*
*       DCC GARCH Forecast        *
*---------------------------------*

Distribution         :  mvnorm
Model                :  DCC(1,1)
Horizon              :  1
Roll Steps           :  0
-----------------------------------

0-roll forecast: 
, , 1

       [,1]   [,2]   [,3]
[1,] 1.0000 0.7472 0.6790
[2,] 0.7472 1.0000 0.6897
[3,] 0.6790 0.6897 1.0000

How do I extract or save the forecast matrix as object so I can use the covariance matrix for portfolio optimization?

Thank you.

Solution

  • The dcc.focast object you create is an S4 object. By default, print calls its show method, which just presents a summary.

    You can access each of the elements in the object using the S4 method calls described in ?"DCCforecast-class" (Note the double quotes). I think the one you want is rcov(dcc.focast), though it could be rcor.

    You can also access the multivariate forecast list using the slot @mforecast: dcc.focast@mforecast and you can get the subsets by adding brackets: dcc.focast@mforecast[1].