This is the mail archive of the gsl-discuss@sources.redhat.com mailing list for the GSL project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: gsl-chisquare cumulative density function


Gilles Frising writes:
 >  I wanted to use the cumulative chisquare pdf of gsl:
 > `gsl_ran_chisq_pdf'.  I checked this function in maple,cernlib and
 > gsl, and i found different result for gsl.
 >
 >  MAPLE:
 >  statevalf[cdf,chisquare[6]](3.);
 >       .1911531695
 >
 >  CERNLIB:
 >  prob: 0.191153
 >
 >  GSL:
 >  gsl_ran_chisq_pdf(3.,6.): 0.125511
 >                            =========
 >
 >  Is this a bug or have i done sth. wrong?

Hi,

The pdf is the probability distribution function (point value), so
this is a different quantity from the cdf. 

The cumulative cdf's are not implemented in the present version of
GSL.  We are looking for a volunteer to write them.

In the meantime you'd have to look up the formula and use the
corresponding routines from the specfunc directory -- it's an
incomplete gamma function, I believe.

regards
Brian Gough


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]