Bayesian Econometric Methods

by Gary Koop , Dale J. Poirier , Justin L. Tobias

• ISBN13: 9780521671736
• ISBN10: 0521671736
• Format: Paperback
• Copyright: 2007-01-15
• Publisher: Cambridge University Press

Summary

This book is a volume in the Econometric Exercises series. It teaches principles of Bayesian econometrics by posing a series of theoretical and applied questions, and providing detailed solutions to those questions. The text is primarily suitable for graduate study in econometrics, though it can be used for advanced undergraduate courses, and should generate interest from students in related fields, including finance, marketing, agricultural economics, business economics, and other disciplines that employ statistical methods. The book provides a detailed treatment of a wide array of models commonly employed by economists and statisticians, including linear regression-based models, hierarchical models, latent variable models, mixture models, and time series models. Basics of random variable generation and simulation via Markov Chain Monte Carlo (MCMC) methods are also provided. Finally, posterior simulators for each type of model are rigorously derived, and MATLAB computer programs for fitting these models (using both actual and generated data sets) are provided on the Web site accompanying the text. Book jacket.

Author Biography

Gary Koop is Professor of Economics at the University of Strathclyde Dale J. Poirier is Professor of Economics at the University of California, Irvine Justin L. Tobias is Associate Professor of Economics, Iowa State University

Table of Contents

List of exercises	p. ix
Preface to the series	p. xv
Preface	p. xix
The subjective interpretation of probability	p. 1
Bayesian inference	p. 11
Point estimation	p. 29
Frequentist properties of Bayesian estimators	p. 37
Interval estimation	p. 51
Hypothesis testing	p. 59
Prediction	p. 71
Choice of prior	p. 79
Asymptotic Bayes	p. 91
The linear regression model	p. 107
Basics of Bayesian computation	p. 117
Monte Carlo integration	p. 119
Importance sampling	p. 124
Gibbs sampling and the Metropolis-Hastings algorithm	p. 128
Other (noniterative) methods for generating random variates	p. 157
Hierarchical models	p. 169
The linear regression model with general covariance matrix	p. 191
Latent variable models	p. 203
Mixture models	p. 253
Some scale mixture of normals models	p. 254
Other continuous and finite-mixture models	p. 260
Bayesian model averaging and selection	p. 281
Bayesian model averaging	p. 282
Bayesian variable selection and marginal likelihood calculation	p. 287
Some stationary time series models	p. 297
Some nonstationary time series models	p. 319
Appendix	p. 335
Bibliography	p. 343
Index	p. 353
Table of Contents provided by Ingram. All Rights Reserved.