Skip to content

pointers on using VaR.gpd with return series?

4 messages · Sylvain BARTHELEMY, Krishna Kumar, Brian G. Peterson

Original
Brian G. Peterson
# 
Does anyone have any hints on utilizing VaR.gpd on return series instead 
of price series?

I have tried converting a return series to a wealth index (using 
cumprod), but this still seems to cause problems with the VaR package.

for example:

library(PerformanceAnalytics)
data(edhec)
wi<-cumprod.column(1+edhec)
library(VaR)
vt<-VaR.gpd(wi[,1])

Error in optim(init, gpd.liklhd, hessian = TRUE, method = "Nelder-Mead") :
         non-finite finite-difference value [2]
In addition: There were 50 or more warnings (use warnings() to see the 
first 50)

A message to the maintainer of the package has without response.

If I don't get any usable feedback, I'll probably move to using fit.GPD 
from QRMlib (which I have had some luck with in the past) to add a 
general Pareto method to the VaR functions in PerformanceAnalytics, as I 
feel that parametric VaR functions on a broader set of distributions and 
copulae should be more widely available.

Regards,

    - Brian
Sylvain BARTHELEMY
# 
I think that there is a problem with the VaR.gpd function, as it works on
USD/EUR and not on EUR/USD values

library(PerformanceAnalytics)
eurusd <- get.hist.quote("EUR/USD", provider="oanda", start = "2006-01-01")
usdeur <- get.hist.quote("USD/EUR", provider="oanda", start = "2006-01-01")

library(VaR)
v1 <- VaR.gpd(as.vector(eurusd))
v2 <- VaR.gpd(as.vector(usdeur))


output:
Error in optim(init, gpd.liklhd, hessian = TRUE, method = "Nelder-Mead") :
         non-finite finite-difference value [2]
In addition: There were 50 or more warnings (use warnings() to see the 
first 50)
There were 11 warnings (use warnings() to see)

---
Sylvain Barth?l?my
Research Director, TAC
www.tac-financial.com | www.sylbarth.com


-----Message d'origine-----
De?: r-sig-finance-bounces at stat.math.ethz.ch
[mailto:r-sig-finance-bounces at stat.math.ethz.ch] De la part de Brian G.
Peterson
Envoy??: mardi 21 ao?t 2007 14:38
??: R-SIG-Finance
Objet?: [R-SIG-Finance] pointers on using VaR.gpd with return series?

Does anyone have any hints on utilizing VaR.gpd on return series instead 
of price series?

I have tried converting a return series to a wealth index (using 
cumprod), but this still seems to cause problems with the VaR package.

for example:

library(PerformanceAnalytics)
data(edhec)
wi<-cumprod.column(1+edhec)
library(VaR)
vt<-VaR.gpd(wi[,1])

Error in optim(init, gpd.liklhd, hessian = TRUE, method = "Nelder-Mead") :
         non-finite finite-difference value [2]
In addition: There were 50 or more warnings (use warnings() to see the 
first 50)

A message to the maintainer of the package has without response.

If I don't get any usable feedback, I'll probably move to using fit.GPD 
from QRMlib (which I have had some luck with in the past) to add a 
general Pareto method to the VaR functions in PerformanceAnalytics, as I 
feel that parametric VaR functions on a broader set of distributions and 
copulae should be more widely available.

Regards,

    - Brian

_______________________________________________
R-SIG-Finance at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. 
-- If you want to post, subscribe first.
Krishna Kumar
#
Sylvain BARTHELEMY wrote:
Ouch the default parameters there are two possible fixes setting the 
cut-off threshold using p.tr helps.

(a)  doing the following call on your data comes back with some results..

 > v1 <- VaR.gpd(as.vector(eurusd),p.tr=0.95)

(b) The other alternate is to rewrite VaR.gpd and set  hessian=FALSE 
where it makes the call to maximize the log-likelihood.

optim(init, gpd.liklhd, *hessian = TRUE*, method = "Nelder-Mead") :


Neither of these are "the solution"  as this is more an Art than a 
science. Method (a) relates to the question of  how to pick the 
threshold. Very few and
 you have biased fit and too many you are no longer fitting the tail.

In this context I would point you towards the evir library and the 
excellent book by Alex Mcneill on this but doing the following should 
give you some hints..

 >require(evir)
 >shape(danish)

Hope this helps,

Best
Krishna
Brian G. Peterson
#
Krishna Kumar wrote:
Kris and Sylvain,

Thanks for the pointers.  I would have assumed that a VaR function would 
set some reasonable defaults for threshold and p value, but I guess not. 
  Basically, I *know* that a GPD distribution is fitable in a reasonable 
fashion to this data, as I've done it.  I was hoping to cut down on my 
implementation difficulties by using functions already written.

I'll try fitting with evir and QRMlib. (the QRMlib package is the R port 
for functions from McNiel's book, which is well worth owning)

Stay tuned...  as always, we'll share our results.

Regards,

   - Brian