Thursday, October 25, 2012

Histogram widths: Bayesian blocks

One of frustrating things is getting the histogram widths right: it has always been an arbitrary procedure, which can be misleading. Here is a astroML implementation of a rigorous procedure to determine the fixed or flexible width histogram bars.
The utility of the Bayesian blocks approach goes beyond simple data representation, however: the bins can be shown to be optimal in a quantitative sense, meaning that the histogram becomes a powerful statistical measure.

No comments:

Post a Comment