Dear Sara,
You could also use the Subbotools package by Giulio Bottazzi and test the
goodness of fit of your data to this distribution.
See:
http://www.lem.sssup.it/WPLem/files/2004-14.pdf
Also, http://cafim.sssup.it/~giulio/software/subbotools/install_cygwin.html
Regards,
José
Mr José Luis
You have to problems:
a) You need the cumulative distribution function, hence I suggest to
look at package "normalp" which offers function pnormp() (as well as
dnormp and other related functions that might be of interest)
b) The KS test won't have much power given you are estimating the
parame
Hi,
I have a Data Set x and I want to check with a Kolmogorow-Smirnow-Test, if x
comes from a Subbotin Distribution, whose density function is:
function(x,location,scale,tail) # Exponential power (=Subbotin)
{
const<- 2*scale*tail^(1/tail) *gamma(1+1/tail)
z<- (x-location)/scale
exp(-1/tai
3 matches
Mail list logo