This thesis explores advanced Bayesian statistical methods for extracting key information for cosmological model selection, parameter inference and forecasting from astrophysical observations. Bayesian model selection provides a measure of how good models in a set are relative to each other - but what if the best model is missing and not included in the set? Bayesian Doubt is an approach which addresses this problem and seeks to deliver an absolute rather than a relative measure of how good a model is. Supernovae type Ia were the first astrophysical observations to indicate the late time acceleration of the Universe - this work presents a detailed Bayesian Hierarchical Model to infer the cosmological parameters (in particular dark energy) from observations of these supernovae type Ia.
The amount of cosmological data has dramatically increased in the past decades due to an unprecedented development of telescopes, detectors and satellites. Efficiently handling and analysing new data of the order of terabytes per day requires not only computer power to be processed but also the development of sophisticated algorithms and pipelines. Aiming at students and researchers the lecture notes in this volume explain in pedagogical manner the best techniques used to extract information from cosmological data, as well as reliable methods that should help us improve our view of the universe.
The three neutrinos are ghostly elementary particles that exist all across the Universe. Though every second billions of them fly through us, they are extremely hard to detect. We used to think they had no mass, but recently discovered that in fact they have a tiny mass. The quest for the neutrino mass scale and mass ordering (specifying how the three masses are distributed) is an extremely exciting one, and will open the door towards new physics operating at energy scales we can only ever dream of reaching on Earth. This thesis explores the use of measurements of the Cosmic Microwave Background (the oldest light reaching us, a snapshot of the infant Universe) and maps of millions of galaxies to go after the neutrino mass scale and mass ordering. Neutrinos might teach us something about the mysterious dark energy powering the accelerated expansion of the Universe, or about cosmic inflation, which seeded the initial conditions for the Universe. Though extremely baffling, neutrinos are also an exceptionally exciting area of research, and cosmological observations promise to reveal a great deal about these elusive particles in the coming years.
Provides an accessible foundation to Bayesian analysis usingreal world models This book aims to present an introduction to Bayesian modellingand computation, by considering real case studies drawn fromdiverse fields spanning ecology, health, genetics and finance. Eachchapter comprises a description of the problem, the correspondingmodel, the computational method, results and inferences as well asthe issues that arise in the implementation of theseapproaches. Case Studies in Bayesian Statistical Modelling andAnalysis: Illustrates how to do Bayesian analysis in a clear and concisemanner using real-world problems. Each chapter focuses on a real-world problem and describes theway in which the problem may be analysed using Bayesianmethods. Features approaches that can be used in a wide area ofapplication, such as, health, the environment, genetics,information science, medicine, biology, industry and remotesensing. Case Studies in Bayesian Statistical Modelling andAnalysis is aimed at statisticians, researchers andpractitioners who have some expertise in statistical modelling andanalysis, and some understanding of the basics of Bayesianstatistics, but little experience in its application. Graduatestudents of statistics and biostatistics will also find this bookbeneficial.
"As telescopes, detectors, and computers grow ever more powerful, the volume of data at the disposal of astronomers and astrophysicists will enter the petabyte domain, providing accurate measurements for billions of celestial objects. This book provides a comprehensive and accessible introduction to the cutting-edge statistical methods needed to efficiently analyze complex data sets from astronomical surveys such as the Panoramic Survey Telescope and Rapid Response System, the Dark Energy Survey, and the upcoming Large Synoptic Survey Telescope. It serves as a practical handbook for graduate students and advanced undergraduates in physics and astronomy, and as an indispensable reference for researchers. The updates in this new edition will include fixing "code rot," correcting errata, and adding some new sections. In particular, the new sections include new material on deep learning methods, hierarchical Bayes modeling, and approximate Bayesian computation. Statistics, Data Mining, and Machine Learning in Astronomy presents a wealth of practical analysis problems, evaluates techniques for solving them, and explains how to use various approaches for different types and sizes of data sets. For all applications described in the book, Python code and example data sets are provided. The supporting data sets have been carefully selected from contemporary astronomical surveys (for example, the Sloan Digital Sky Survey) and are easy to download and use. The accompanying Python code is publicly available, well documented, and follows uniform coding standards. Together, the data sets and code enable readers to reproduce all the figures and examples, evaluate the methods, and adapt them to their own fields of interest"--
Astrostatistical Challenges for the New Astronomy presents a collection of monographs authored by several of the disciplines leading astrostatisticians, i.e. by researchers from the fields of statistics and astronomy-astrophysics, who work in the statistical analysis of astronomical and cosmological data. Eight of the ten monographs are enhancements of presentations given by the authors as invited or special topics in astrostatistics papers at the ISI World Statistics Congress (2011, Dublin, Ireland). The opening chapter, by the editor, was adapted from an invited seminar given at Los Alamos National Laboratory (2011) on the history and current state of the discipline; the second chapter by Thomas Loredo was adapted from his invited presentation at the Statistical Challenges in Modern Astronomy V conference (2011, Pennsylvania State University), presenting insights regarding frequentist and Bayesian methods of estimation in astrostatistical analysis. The remaining monographs are research papers discussing various topics in astrostatistics. The monographs provide the reader with an excellent overview of the current state astrostatistical research, and offer guidelines as to subjects of future research. Lead authors for each chapter respectively include Joseph M. Hilbe (Jet Propulsion Laboratory and Arizona State Univ); Thomas J. Loredo (Dept of Astronomy, Cornell Univ); Stefano Andreon (INAF-Osservatorio Astronomico di Brera, Italy); Martin Kunz ( Institute for Theoretical Physics, Univ of Geneva, Switz); Benjamin Wandel ( Institut d'Astrophysique de Paris, Univ Pierre et Marie Curie, France); Roberto Trotta (Astrophysics Group, Dept of Physics, Imperial College London, UK); Phillip Gregory (Dept of Astronomy, Univ of British Columbia, Canada); Marc Henrion (Dept of Mathematics, Imperial College, London, UK); Asis Kumar Chattopadhyay (Dept of Statistics, Univ of Calcutta, India); Marisa March (Astrophysics Group, Dept of Physics, Imperial College, London, UK)./body
Digital sky surveys, high-precision astrometry from satellite data, deep-space data from orbiting telescopes, and the like have all increased the quantity and quality of astronomical data by orders of magnitude per year for several years. Making sense of this wealth of data requires sophisticated statistical techniques. Fortunately, statistical methodologies have similarly made great strides in recent years. Powerful synergies thus emerge when astronomers and statisticians join in examining astrostatistical problems and approaches. The book begins with an historical overview and tutorial articles on basic cosmology for statisticians and the principles of Bayesian analysis for astronomers. As in earlier volumes in this series, research contributions discussing topics in one field are joined with commentary from scholars in the other. Thus, for example, an overview of Bayesian methods for Poissonian data is joined by discussions of planning astronomical observations with optimal efficiency and nested models to deal with instrumental effects. The principal theme for the volume is the statistical methods needed to model fundamental characteristics of the early universe on its largest scales.
This volume contains a selection of chapters based on papers to be presented at the Fifth Statistical Challenges in Modern Astronomy Symposium. The symposium will be held June 13-15th at Penn State University. Modern astronomical research faces a vast range of statistical issues which have spawned a revival in methodological activity among astronomers. The Statistical Challenges in Modern Astronomy V conference will bring astronomers and statisticians together to discuss methodological issues of common interest. Time series analysis, image analysis, Bayesian methods, Poisson processes, nonlinear regression, maximum likelihood, multivariate classification, and wavelet and multiscale analyses are all important themes to be covered in detail. Many problems will be introduced at the conference in the context of large-scale astronomical projects including LIGO, AXAF, XTE, Hipparcos, and digitized sky surveys.
In 1978 Edwin T. Jaynes and Myron Tribus initiated a series of workshops to exchange ideas and recent developments in technical aspects and applications of Bayesian probability theory. The first workshop was held at the University of Wyoming in 1981 organized by C.R. Smith and W.T. Grandy. Due to its success, the workshop was held annually during the last 18 years. Over the years, the emphasis of the workshop shifted gradually from fundamental concepts of Bayesian probability theory to increasingly realistic and challenging applications. The 18th international workshop on Maximum Entropy and Bayesian Methods was held in Garching / Munich (Germany) (27-31. July 1998). Opening lectures by G. Larry Bretthorst and by Myron Tribus were dedicated to one of th the pioneers of Bayesian probability theory who died on the 30 of April 1998: Edwin Thompson Jaynes. Jaynes revealed and advocated the correct meaning of 'probability' as the state of knowledge rather than a physical property. This inter pretation allowed him to unravel longstanding mysteries and paradoxes. Bayesian probability theory, "the logic of science" - as E.T. Jaynes called it - provides the framework to make the best possible scientific inference given all available exper imental and theoretical information. We gratefully acknowledge the efforts of Tribus and Bretthorst in commemorating the outstanding contributions of E.T. Jaynes to the development of probability theory.