Bug

Bug

[Przemyslaw Sliwa]

All,

I have found a bug in the routine which computes the cdf of the student t-distrribution.

Passing INF as the argument of gsl_cdf_tdist_Q gives wrong result of -0.001, which is not correct at all.

Thanks

Przemyslaw



--------------------------------------------------
Deutsche Bank AG
Risk Controlling
Credit Risk Engineering & Implementation
            /  Daily Credit Risk

Tel.:  (+44 20) 754 59676 London
Fax:   (+44 20) 754  71060 London

----------------------------------------------------


--

This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.

Re: Bug

[Jason H. Stover]

I was unable to reproduce the bug with this code:

#include <gsl/gsl_math.h>
#include <gsl/gsl_cdf.h>

int
main ()
{
  double x = GSL_POSINF;
  double y;

  y = gsl_cdf_tdist_Q ( x, 15.0);
  printf ("%f %f\n",x,y);
  return;
}

compiled with gcc test.c -lgsl -lm.

-Jason


On Wed, Jul 07, 2004 at 10:37:56AM +0100, Przemyslaw Sliwa wrote:
> All,
> 
> I have found a bug in the routine which computes the cdf of the student t-distrribution.
> 
> Passing INF as the argument of gsl_cdf_tdist_Q gives wrong result of -0.001, which is not correct at all.
> 
> Thanks
> 
> Przemyslaw
> 
> 
> 
> --------------------------------------------------
> Deutsche Bank AG
> Risk Controlling
> Credit Risk Engineering & Implementation
>             /  Daily Credit Risk
> 
> Tel.:  (+44 20) 754 59676 London
> Fax:   (+44 20) 754  71060 London
> 
> ----------------------------------------------------
> 
> 
> --
> 
> This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
> 
> 
> 

Re: Bug

[Brian Gough]

Przemyslaw Sliwa writes:
 > I have found a bug in the routine which computes the cdf of the student t-distrribution.
 > 
 > Passing INF as the argument of gsl_cdf_tdist_Q gives wrong result of -0.001, which is not correct at all.

Please send bug reports to bug-gsl@gnu.org with an example program
that reproduces the problem, plus all the details of the operating
system and compiler.  Thanks.

-- 
Brian Gough

Network Theory Ltd,
Commercial support for GSL --- http://www.network-theory.co.uk/gsl/

bug

[Fernando]

Hi,
I have found a bug in the hypergeometric function routine, F(a,b,x).
When a is negative ( a<-2), the function has a discontinuity.
I have include a program which just plots the function.
In the program is commented at which values of x the discontinuity
begins.

thank you,
Fernando Pilotto