[R] A question about riskmeasures() vs. qgpd() in library(evir)

Jacques Wagnor jacques.wagnor at gmail.com
Wed Jun 6 15:42:54 CEST 2007


Dear List,

This inquiry probably does not directly pertain to R.

I am using library(evir) to learn EVT.  Based on my reading of things,
it is my understanding that if one wants to calculate quantiles of
GPD, one could use either riskmeasures() or qgpd().  However, using
data(danish) as an example, the quantile estimates produced by
riskmeasures() are considerably different from those produced by
qgpd() as follows:

> library(evir)
> data(danish)
> out <- gpd(danish, 10)
> riskmeasures(out, c(0.999, 0.9999))

              p  quantile     sfall
[1,] 0.9990  94.28956   191.3697
[2,] 0.9999  304.62448 609.3696

> qgpd(c(0.999, 0.9999), out$par.ests["xi"], out$par.ests["beta"])

[1]  67.22493 200.41271

Any insights would be greatly appreciated.

Regards,

Jacques

platform       i386-pc-mingw32
arch           i386
os             mingw32
system         i386, mingw32
status
major          2
minor          5.0
year           2007
month          04
day            23
svn rev        41293
language       R
version.string R version 2.5.0 (2007-04-23)



More information about the R-help mailing list