[R-SIG-Finance] how to test significance of VAR coefficients in DCC GARCH Fit

alexios ghalanos alexios at 4dscape.com
Wed Mar 14 17:34:43 CET 2012


Currently the DCCfit object (returned from running dccfit) does not 
return all the information on the VAR (coefficients can be extracted by 
looking at the model slot and 'varcoef' list i.e. fit at model$varcoef).

A better approach is to first estimate the VAR model using the function
'varxfit' in the package which returns the standard errors and all 
relevant information, and then passing this returned object to the 
dccfit routine (example follows).

#################
library(rmgarch)
data(dji30ret)
Data = dji30ret[, 1:3, drop = FALSE]

vfit = varxfit(X=Data, p=1, exogen = NULL, robust = FALSE,
gamma = 0.25, delta = 0.01, nc = 10, ns = 500, postpad = "constant")

uspec = ugarchspec(mean.model = list(armaOrder = c(0,0), include.mean = 
FALSE), variance.model = list(garchOrder = c(1,1), model = "sGARCH"),
distribution.model = "norm")

spec = dccspec(uspec = multispec( replicate(3, uspec) ), VAR = TRUE,
lag = 1, dccOrder = c(1,1), asymmetric = FALSE, distribution = "mvnorm")

fit = dccfit(spec, data = Data, fit.control = list(eval.se=TRUE), 
VAR.fit = vfit)
#################

The package also includes for convenience the 'varxfilter', 
'varxforecast' and 'varxsim' functions which are used by the 
multivariate garch routines internally.

As mentioned in the documentation, a comprehensive list of examples are 
included in the 'inst/rmgarch.tests' folder of the package.

Regards,

Alexios


On 14/03/2012 16:04, mamush bukana wrote:
> Dear all,
> I fit a  DCC-GARCH model using "dccfit" function in "rmgarch" package. The
> "show()" method there gives summary of the GARCH and DCC parameters only,
> but not of the VAR ones. May you suggest me how to get the summary
> (including significance test) of the VAR parameters please?
>
> Many thanks as usual
>
> Mamush
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.
>
>



More information about the R-SIG-Finance mailing list