Göttingen 2012 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
T: Fachverband Teilchenphysik
T 74: Experimentelle Methoden 1
T 74.4: Talk
Monday, February 27, 2012, 17:30–17:45, VG 0.111
Bayesian Analysis Toolkit - BAT — Frederik Beaujean1, Allen Caldwell1, Daniel Kollár1, Kevin Kröninger2, •Shabnaz Pashapour2, and Arnulf Quadt2 — 1Max-Planck-Institute for Physics — 2University of Göttingen
Statistical treatment of data is an essential part of any data analysis and interpretation. Different statistical methods and approaches can be used, however the implementation of these approaches is complicated and at times inefficient.
The Bayesian Analysis Toolkit (BAT) is a software package developed in C++ framework that facilitates the statistical analysis of the data using Bayesian Theorem. The tool evaluates the posterior probability distributions for models and their parameters using Markov Chain Monte Carlo which in turn provide straightforward parameter estimation, limit setting and uncertainty propagation. Additional algorithms, such as Simulated Annealing, allow extraction of the global mode of the posterior.
BAT sets a well-tested environment for flexible model definition and also includes a set of predefined models for standard statistical problems. The package is interfaced to other software packages commonly used in high energy physics, such as ROOT, Minuit, RooStats and CUBA.
We present a general overview of BAT and its algorithms. A few physics examples are shown to introduce the spectrum of its applications. In addition, new developments and features are summarized.