bug in gevMoments(fExtremes)

[Jens Oehlschlägel]

Dear all,

just wanted to let you know that gevMoments() gave me an apparently wrong mean for the gumbel distribution for mu=-EulerConstant. 

Best regards


Dr. Jens Oehlschlägel



> library(fExtremes)

> EulerConstant <-  0.57721566490153286060651209008240243104215933519399235988057672348848677267776646709369470632917467495
> 
> gevMoments(xi=0)
$param
  xi   mu beta 
   0    0    1 

$mean
[1] 0.5772157

$var
[1] 1.644934

> gevMoments(xi=0, mu=-EulerConstant)
$param
        xi         mu       beta 
 0.0000000 -0.5772157  1.0000000 

$mean
[1] -0.5772157

$var
[1] 1.644934


> meanvar <- function(x)c(mean=mean(x), var=var(x))
> meanvar(rgev(1000000, xi=0))
     mean       var 
0.5793065 1.6445646 
> meanvar(rgev(1000000, xi=0, mu=-EulerConstant))
        mean          var 
-0.000443175  1.645556911 


fExtremes Version 260.72

> version
               _                           
platform       i386-pc-mingw32             
arch           i386                        
os             mingw32                     
system         i386, mingw32               
status                                     
major          2                           
minor          6.1                         
year           2007                        
month          11                          
day            26                          
svn rev        43537                       
language       R                           
version.string R version 2.6.1 (2007-11-26)