Re: [Help-gsl] Re: Random number generation

[Carl Boettiger]

The number shouldn't be close to zero.  Divide by 600 and you'll have
the mean, which will be close to zero.  Just sum a bunch of gaussian
random numbers of standard deviation sigma and you expect a
displacement of sqrt(2 sigma T), i.e. sqrt(1200) approx 34 -- you're
simulating Brownian motion.

-Carl

On Tue, May 5, 2009 at 2:10 PM, Rodney Sparapani <address@hidden> wrote:
> Vibhuti Dave wrote:
>>
>> Hello,
>>
>> I have attached the code that I am using to generate random numbers that
>> have a Gaussian distribution with a mean value of 0.
>> Ideally, if I generate these numbers and sum them up, the sum should be
>> very
>> close to 0 since it is a gaussian ditribution. I am using the function
>> gsl_ran_gaussian to generate them. However, the mean value seems to be way
>> off 0 when I run the program.
>> Any idea why that might be happening?
>> Thanks
>> Vibhuti
>
> What happens if you set mu = 0.?
>
>
>
> _______________________________________________
> Help-gsl mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/help-gsl
>



-- 
Carl Boettiger
Population Biology, UC Davis
http://two.ucdavis.edu/~cboettig