[SciPy-User] Question about gaussian_kde
Mon Apr 12 12:48:01 CDT 2010
On Mon, Apr 12, 2010 at 12:00 PM, Jorge Scandaliaris
> I am using gaussian_kde mainly to visualize the distribution of some 2D
> measurements. It works ok but there's something I don't understand. When I
> evaluate the estimated pdf, the peaks have values larger than one. If I use
> integrate_box(), however, the results seem correct. How should I normalize the
> values obtained from evaluate? Dividing by the number of datapoints?
pdf is a density function on continuous support, there is no
requirement that it has any bounded value.
It only needs to integrate to one, and be non-negative
Only the pmf of discrete distributions is required to have values
smaller than one.
> SciPy-User mailing list
More information about the SciPy-User