An update on the goodness-of-fit statistical toolkit

The present project aims to develop an open-source and object-oriented software Toolkit for statistical data analysis. Its statistical testing component (the Goodness-of-Fit Statistical Toolkit) contains a variety of one dimensional Goodness-of-Fit tests, from Chi-squared to Kolmogorov-Smirnov, to less known, but generally much more powerful tests such as Anderson-Darling, Cramèr-von Mises, Kuiper, Watson, … The GoF Statistical Toolkit is open-source and downloadable from the web, with its user and software documentation. The component-based design allowed an extension of the GoF Statistical Toolkit: less known, but generally more powerful GoF tests based on EDF-statistics have been recently added to the toolkit. A much more complete variety of GoF inferences is now offered the user, and “standard” GoF tests have been complemented by more “exotic” ones. The weighted formulations of some GoF tests (Kolmogorov-Smirnov and Cramèr-von Mises) have been implemented. Approximations of the distribution of some of the existing GoF tests to the Chi-squared one (Kolmogorov-Smirnov, Cramèr-von Mises, and Watson approximations) are now available in the GoF Statistical Toolkit. Moreover, a layer for user input from ROOT objects has been easily added recently, thanks to the component-based architecture. We present the recent improvements and extensions of the GoF Statistical Toolkit, describing the new statistics methods implemented, and an outlook towards future developments.