Bayesian Econometric Methods examines principles of Bayesian inference by posing a series of theoretical and applied questions and providing detailed solutions to those questions. This second edition adds extensive coverage of models popular in finance and macroeconomics, including state space and unobserved components models, stochastic volatility models, ARCH, GARCH, and vector autoregressive models. The authors have also added many new exercises related to Gibbs sampling and Markov Chain Monte Carlo (MCMC) methods. The text includes regression-based and hierarchical specifications, models based upon latent variable representations, and mixture and time series specifications. MCMC methods are discussed and illustrated in detail - from introductory applications to those at the current research frontier - and MATLAB (R) computer programs are provided on the website accompanying the text. Suitable for graduate study in economics, the text should also be of interest to students studying statistics, finance, marketing, and agricultural economics.
Industry Reviews
'This volume invigorates the understanding and application of Bayesian econometrics with a uniquely constructive, hands-on approach. By moving seamlessly between theory, methods, and applications, it builds understanding and skills that will serve the novice Bayesian econometrician well, and synthesizes the subject for experienced Bayesian practitioners.' John Geweke, Charles R. Nelson Endowed Professor in Economics, University of Washington
'This book is a terrific resource for anybody who would like to study Bayesian econometrics. It is a thoughtfully crafted textbook in which each chapter contains a brief introduction, followed by carefully chosen learning-by-doing problems with detailed and instructive solutions.' Frank Schorfheide, University of Pennsylvania
'... a valuable companion, introducing foundations and methods of the Bayesian approach; elaborates on building blocks of Bayesian inference, model specification and selection, decision-making, and diagnostics; covers most popular uni- and multivariate modeling, including hierarchical and latent variable models; references original literature, also serving researchers looking for a brief introduction to specific topics.' Sylvia Kaufmann, Study Center Gerzensee, Switzerland
'This is an excellent contribution that will greatly expand the understanding of Bayesian econometric methods. Students and instructors will find the easy-to-follow structure and many clearly developed exercises, which reference several recent advances, will build understanding and lead to new insights and better approaches to analysis.' Rodney Strachan, University of Queensland
'This is a wonderful coverage of Bayesian econometrics: from its underlying principles to details of its numerical implementation, all in the context of the key models used in empirical analysis. It will be an invaluable resource for students and researchers alike, and I cannot recommend it too highly.' Gael Martin, Monash University, Australia
'This is an excellent introductory textbook of Bayesian econometrics for senior undergraduate students and graduate students. Unlike other typical textbooks, it nicely illustrates mathematical derivations in detail as solutions of many exercises. Moreover, Matlab computer programs on the website will help understanding of recent simulation methods such as Markov chain Monte Carlo.' Yasuhiro Omori, University of Tokyo
'The text offers broad, thorough, and accessible coverage of important topics in Bayesian econometrics. Delivering both a solid treatment of the foundations of inference and an extensive survey of methodology, and models that are illustrated with numerous empirical examples, the book is an invaluable resource for the practitioner.' Ivan Jeliazkov, University of California, Irvine
'This is a clear, concise, and, above all, practical introduction to Bayesian econometrics. Graduate and advanced undergraduate students will find here a self-contained introduction to Bayesian theory, computation, and applied econometric modeling that can accompany them well into their studies.' William J. McCausland, Universite de Montreal
