This practical guide covers the essential tasks in statistical data analysis encountered in high energy physics and provides comprehensive advice for typical questions and problems. The basic methods for inferring results from data are presented as well as tools for advanced tasks such as improving the signal-to-background ratio, correcting detector effects, determining systematics and many others. Concrete applications are discussed in analysis walkthroughs. Each chapter is supplemented by numerous examples and exercises and by a list of literature and relevant links. The book targets a broad readership at all career levels - from students to senior researchers. An accompanying website provides more algorithms as well as up-to-date information and links. * Free solutions manual available for lecturers at www.wiley-vch.de/supplements/

Data Analysis Techniques for High Energy Physics Experiments

Modern analysis of HEP data needs advanced statistical tools to separate signal from background. This is the first book which focuses on machine learning techniques. It will be of interest to almost every high energy physicist, and, due to its coverage, suitable for students.

This concise set of course-based notes provides the reader with the main concepts and tools needed to perform statistical analyses of experimental data, in particular in the field of high-energy physics (HEP). First, the book provides an introduction to probability theory and basic statistics, mainly intended as a refresher from readers’ advanced undergraduate studies, but also to help them clearly distinguish between the Frequentist and Bayesian approaches and interpretations in subsequent applications. More advanced concepts and applications are gradually introduced, culminating in the chapter on both discoveries and upper limits, as many applications in HEP concern hypothesis testing, where the main goal is often to provide better and better limits so as to eventually be able to distinguish between competing hypotheses, or to rule out some of them altogether. Many worked-out examples will help newcomers to the field and graduate students alike understand the pitfalls involved in applying theoretical concepts to actual data. This new second edition significantly expands on the original material, with more background content (e.g. the Markov Chain Monte Carlo method, best linear unbiased estimator), applications (unfolding and regularization procedures, control regions and simultaneous fits, machine learning concepts) and examples (e.g. look-elsewhere effect calculation).

Advanced Computing and Analysis Techniques in Physics Research

by 2001-09-21

Author: International Workshop on Advanced Computing and Analysis Techniques in Phyics Research

Over the next decade or two, an impressive array of scientific instruments at the Tevatron, RHIC (Relativistic Heavy Ion Collider) and LHC (Large Hadron collider), LIGO (Laser Interferometer Gravitational Observatory) and SDSS (Sloan Digital Sky Survey), to name a few, will usher in the most comprehensive program of study of the fundamental forces of nature and the structure of the universe. Major discoveries are anticipated. But, it is our conviction that the pace of discoveries will be severely impeded unless a concerted effort is made to deploy and employ advanced computing techniques to handle, process and analyze the unprecedented amounts of data. The workshop followed four main tracks: Artificial Intelligence (neural networks and other adaptive multivariate methods); Innovative Software Algorithms and Tools; Symbolic Problem Solving; and Very Large Scale Computing. The workshop covered applications in high energy physics, astrophysics, accelerator physics and nuclear physics. Topics included are: uses of C++ in scientific computing, large scale simulations, advanced analysis environments, worldwide computing; artificial intelligence: online application of neural networks, applications in data analysis, theoretical aspects innovative software algorithms and tools: online monitoring and controls, physics analysis and reconstruction algorithms, pattern recognition techniques, common libraries, grid and distributed computing techniques; symbolic problem solving: Freynman diagram algorithms and tools, symbolic manipulation via function objects, symbolic techniques for Feynman diagrams, multi-loop calculations and results. very large scale computing: online monitoring and controls, analysis farms and DAQ systems, grid architectures

Over the next decade or two, an impressive array of scientific instruments at the Tevatron, RHIC (Relativistic Heavy Ion Collider) and LHC (Large Hadron collider), LIGO (Laser Interferometer Gravitational Observatory) and SDSS (Sloan Digital Sky Survey), to name a few, will usher in the most comprehensive program of study of the fundamental forces of nature and the structure of the universe. Major discoveries are anticipated. But, it is our conviction that the pace of discoveries will be severely impeded unless a concerted effort is made to deploy and employ advanced computing techniques to handle, process and analyze the unprecedented amounts of data. The workshop followed four main tracks: Artificial Intelligence (neural networks and other adaptive multivariate methods); Innovative Software Algorithms and Tools; Symbolic Problem Solving; and Very Large Scale Computing. The workshop covered applications in high energy physics, astrophysics, accelerator physics and nuclear physics. Topics included are: uses of C++ in scientific computing, large scale simulations, advanced analysis environments, worldwide computing; artificial intelligence: online application of neural networks, applications in data analysis, theoretical aspects innovative software algorithms and tools: online monitoring and controls, physics analysis and reconstruction algorithms, pattern recognition techniques, common libraries, grid and distributed computing techniques; symbolic problem solving: Freynman diagram algorithms and tools, symbolic manipulation via function objects, symbolic techniques for Feynman diagrams, multi-loop calculations and results. very large scale computing: online monitoring and controls, analysis farms and DAQ systems, grid architectures