[SciPy-User] visual inspection of kernel density estimators in 2d

[josef.pktd at gmail.com]

I'm trying to find a visual way to see whether a kernel density
estimator "looks" good in 2d. For univariate it is easy to compare
histogram and the estimated density.

Does anyone know what graphs would give a good visual inspection?

My attempt with contours, which should show some oversmoothing if the
default gaussian_kde is used on mixture distributions:

http://picasaweb.google.com/josef.pktd/Joepy#5611119163809333922

(To avoid spamming the list with graphs, I activated a picasa account.)

Josef