rmgarch: Alexios Ghalanos cordist/nisurface/copulagarch

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

rmgarch: Alexios Ghalanos cordist/nisurface/copulagarch

prof.amit.mittal
Hi,

 

I struck out with the following queries , mostly pertaining to graphing in
rmgarch

 

1. Cordist()

 

> cordist(imfdcc.fit@mfit$R, distance = "eigen", n=100, plot = TRUE, title =
"DCC fit rolling Correlations plot", dates = NULL)

Error in seq.default(1, T, by = n) : 'to' must be of length

 

Dccfit objects adjusted for any length of xts object are unable to get a
response apart from above error

 

I tried xts length according to chosen n but T by n is not satisfied.

 

2. Nisurface()

 

Nisurface() seems to give out very similar looking plots for any and every
data in the contour option. The labeling in the surface option differs
according to the chosen pair=c(n,m)  

 

3. For Cgarchfit() result objects, I have no plot options? Cannot find
any in the package. Has anyone used it to show output apart from listing
coefs.

 

 

I am using 13 assets. In the final case, cGarchfit() , I usually have
n(n-1)/2 or 78 copula parameters and none in time varying options when the
model just reports dcc parameters making it impossible to compare. The rest
is just univariate garch.

 

The plots from new rmgarch 1_3_5 package made my life shine with all 78
plots of exceedingly high quality and I hope to join more of them together a
s long data.frame in ggplot2()

 

 


        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] 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.