Likelihood between TH1VsTH1 or TH1VsTF1?

ROOT
1

Hello,

In some cases (e.g. very low stat, or physics prediction), Chi2 test between two histos is not good enough and a likelihood comparison is needed.

So my question is :
Is there an easy way to compute the likelihood between two histograms? or between an histogram and a TF1?

Thanks for help and advices
Loic

2

Hello,

a likelihood cannot be used for a goodness of fit test. If you need to perform a comparison between data (histogram) or between the data and a function prediction, it is better to use the Kolmogorov-Smirnov test in case of low statistics.
See root.cern.ch/root/htmldoc/TH1.ht … ogorovTest

Lorenzo