Determining optimal number of bins for chi squared distribution

Hello,

I am applying a chi squared distribution to histogram. Is there any way in matlab to determine the optimal number of bins to get the lowest chi squared (X2) statistic?

The Chi-square goodness-of-fit test function does not necessarily tell you the optimal number of bins to use, or does it?

Many thanks

John

Answer 1

[Damiano Capocci]

More you have, better will be. if you have a great number of data you can go far with stand deviations because you will always have the product Prob*N_tot big enough