Dear all,
I'm using R 2.8.1 on Windows XP.
I want to loop over several garch-models to evaluate the best model-fit.
After loading the package
fGarch
I execute the following loop:
For(i in 1:5){
for(j in 1:5){
garchFit(~ arma(0,0) + garch(i,j), stetige_renditen[,7],
cond.dist="sged", trace=F)
}
}
Where "stetige_renditen[,7]" are the continously compounded returns of
the FX €/$ 2007 - 2008.
Unfortunately I get the following error:
[1] "data" "i" "j"
[1] "data"
Fehler in .garchArgsParser(formula = formula, data = data, trace = FALSE) :
Formula and data units do not match.
On top the time series is not stationary. So any advice to choose another model?
I think IGARCH is not so good...
Any help about the code and the non-statioarity-thing would be very appreciated.
Kind regards,
Thomas
[[alternative HTML version deleted]]