So if I understand what you are trying to do..you are doing something like
the following:
X = randn(200);
det(cov(X))
And then what this returns is a number very close to (or equal to) 0.
Right?
And I guess your logic is that since X is iid Gaussian (of say var 1), you
expect cov_X to be identity
matrix and so det should be 200, not 0.
Is that the problem?
Nandan
On 8/2/07, PRAJIT S NAIR wrote:
>
> Dear Roshan
>
> Try format command
> >> help format
>
> With Warm Wishes
> Prajit S Nair
>
> On 8/2/07, Roshan Martis < r...@yahoo.com> wrote:
> >
> > Dera Sir,
> > I am new to the area of statistics. I am trying to find the determinant
> > of a 200*200 covariance matrix of a Gaussian distribution. Due to so many
> > multiplications of fractional numbers it gives O as the result. Can
anybody
> > tell me how to solve this particular problem.
> >
> > Thanking you
> >
> > ROSHAN JOY MARTIS
> >
> >