Skip to content

PerformanceAnalytics: Error computing component-VaR using method="kernel"

2 messages · Gero Schwenk, Brian G. Peterson

#
Hi there and a happy new year!
I've got a question regarding the VaR-function from the 
PerformanceAnalytics-package. When I call it in order to compute VaR 
contributions using the nonparametric kernel estimator, I get an error 
report. Obviously this is independent of my data, because I can 
replicate it using the edhec example data, as shown below. Does anybody 
have a hint how to deal with this problem?

Cheers,
Gero

###

library(PerformanceAnalytics)
data(edhec)
VaR(R=edhec, method="kernel", portfolio_method = "component")

###

 > VaR(R=edhec, method="kernel", portfolio_method = "component")
no weights passed in, assuming equal weighted portfolio
Fehler in `colnames<-`(`*tmp*`, value = c("Convertible Arbitrage", "CTA 
Global",  :
  attempt to set colnames on object with less than two dimensions
#
On 01/01/2011 11:23 AM, Gero Schwenk wrote:
Looks like a bug, I can replicate it here.  Not sure what's broken yet, 
as this all had to work when we sent it to CRAN recently.  We'll debug 
and patch.  ES is broken too with method='kernel'

Regards,

   - Brian