BAT - Bayesian Analysis Toolkit

The Bayesian Analysis Toolkit is a software analysis package which allows to address the main goals of a typical data analysis:

  • compare model predictions with data,
  • draw conclusions on the validity of the model as a representation of the data and
  • extract the possible values of parameters within the context of a model.

The BAT is based on Bayes' Theorem and is realized with the use of Markov Chain Monte Carlo. This gives access to the full posterior probability distribution and enables straightforward parameter estimation, limit setting and uncertainty propagation.

The BAT is implemented in C++ on top of CERN's ROOT with emphasis put on flexibility and modularity in defining models while keeping in mind the reliability and speed requirements of the numerical operations.

Authors

See doc/CREDITS file for list of contributors to BAT.

Availability

All BAT releases are available from http://mpp.mpg.de/bat/. The source code is managed online at https://github.com/bat/bat. BAT is open-source software under the LGPLv3 or later. See doc/COPYING and doc/LICENSE for licensing details.

Contents

Directory bat/:

  • doc/: documentation about BAT
  • examples/: well commented example programs, see examples/README.md
  • BAT/: BAT header files
  • src/: source files of the BAT core library
  • models: models for specific data-analysis problems: fast fitters in base/ and the multi-template fitter in mtf
  • benchmarks: performance test suite with html output
  • test: unit tests
  • tools: template script to generate a BAT project and build utilities
  • INSTALL.md: instructions to install BAT on your system
  • README.md: this document

Other files distributed with BAT are part of the configuration and build system.

Installation

See the INSTALL.md file for installation instructions. The instructions for the latest development version are also available at https://github.com/bat/bat/blob/master/INSTALL.md.

Contact

For additional information and contacting the authors, please consult the BAT web page at http://mpp.mpg.de/bat/. If you want to report an error or file a request, please file an issue at https://github.com/bat/bat/issues.