Logo ROOT  
Reference Guide
Goodness of Fit Tests

Classical one-dimensional goodness of git tests for unbinned data.

ROOT provides 1 sample goodness of fit test (comparison of data with a theoretical distribution) and 2-sample test (comparison of two data sets) through the class ROOT::Math::GoFTest The algorithms provided are the Kolmogorov-Smirnov and Anderson-Darling. These tests could be applied approximately also to binned data, assuming the bin size is much smaller than the intrinsic data variations. It is assumed than a bin is like many data at the same bin center value. For these binned version tests look at TH1::KolmogorovTest and TH1::AndersonDarlingTest


class  ROOT::Math::GoFTest
 GoFTest class implementing the 1 sample and 2 sample goodness of fit tests for uni-variate distributions and data. More...