For the impatient
For a quick tutorial covering the basic features, jump directly to the basic tutorial.
The purpose of this document is to introduce the Bayesian Analysis Toolkit (BAT), a C++ package providing tools to
- compare model predictions with data,
- draw conclusions on the validity of a model as a representation of the data,
- and to extract the values of the free parameters of a model.
BAT was originally developed at the Max Planck institute for physics in Munich, Germany, in the context of particle physics. BAT uses ROOT for data handling and graphical output.
How to read this document
If you are new to BAT, we recommend to start with the [basic tutorial](Getting started) to quickly get you going. If you then want to learn more, check the brief introduction to the statistical basics in Bayesian Statistics. The core algorithm of BAT is Markov chain Monte Carlo, described in Markov chain Monte Carlo. Further chapters describe other tools of BAT, such as Optimization, specific statistical models, and advanced features.
Further help
Please visit our home page for an overview of papers, research, and other activities related to BAT. The code development takes place over at github.
This document is meant to is quickly enable to you to find your way around BAT. We describe the common workflow and explain how to solve some not-so-common tasks. To keep the size of this document manageable, we refer to the html reference guide in which all classes, methods etc. are documented. The BAT source code comes with many examples that illustrate how to use the various features. You can browse them online.
If you run into a problem with BAT, the preferred method is to create a new issue on github. This allows other users to easily find your issue online and benefit from the solution. If you provide some code for us to reproduce the problem, we can help you much faster. Have a look at issues with the label troubleshooting or question.
For general questions or comments that do not belong into the public in the form of issues, you can contact the developers at bat@mpp.mpg.de.