This book introduces a new generation of statistical econometrics. After linear models leading to analytical expressions for estimators, and non-linear models using numerical optimization algorithms, the availability of high- speed computing has enabled econometricians to consider econometric models without simple analytical expressions. The previous difficulties presented by the presence of integrals of large dimensions in the probability density functions or in the moments can be circumvented by a simulation-based approach. After a brief survey of classical parametric and semi-parametric non-linear estimation methods and a description of problems in which criterion functions contain integrals, the authors present a general form of the model where it is possible to simulate the observations. They then move to calibration problems and the simulated analogue of the method of moments, before considering simulated versions of maximum likelihood, pseudo-maximum likelihood, or non-linear least squares. The general principle of indirect inference is presented and is then applied to limited dependent variable models and to financial series.

This substantial volume has two principal objectives. First it provides an overview of the statistical foundations of Simulation-based inference. This includes the summary and synthesis of the many concepts and results extant in the theoretical literature, the different classes of problems and estimators, the asymptotic properties of these estimators, as well as descriptions of the different simulators in use. Second, the volume provides empirical and operational examples of SBI methods. Often what is missing, even in existing applied papers, are operational issues. Which simulator works best for which problem and why? This volume will explicitly address the important numerical and computational issues in SBI which are not covered comprehensively in the existing literature. Examples of such issues are: comparisons with existing tractable methods, number of replications needed for robust results, choice of instruments, simulation noise and bias as well as efficiency loss in practice.

This book describes the new generation of discrete choice methods, focusing on the many advances that are made possible by simulation. Researchers use these statistical methods to examine the choices that consumers, households, firms, and other agents make. Each of the major models is covered: logit, generalized extreme value, or GEV (including nested and cross-nested logits), probit, and mixed logit, plus a variety of specifications that build on these basics. Simulation-assisted estimation procedures are investigated and compared, including maximum stimulated likelihood, method of simulated moments, and method of simulated scores. Procedures for drawing from densities are described, including variance reduction techniques such as anithetics and Halton draws. Recent advances in Bayesian procedures are explored, including the use of the Metropolis-Hastings algorithm and its variant Gibbs sampling. The second edition adds chapters on endogeneity and expectation-maximization (EM) algorithms. No other book incorporates all these fields, which have arisen in the past 25 years. The procedures are applicable in many fields, including energy, transportation, environmental studies, health, labor, and marketing.

Monte Carlo Simulation for Econometricians presents the fundamentals of Monte Carlo simulation (MCS), pointing to opportunities not often utilized in current practice, especially with regards to designing their general setup, controlling their accuracy, recognizing their shortcomings, and presenting their results in a coherent way. The author explores the properties of classic econometric inference techniques by simulation. The first three chapters focus on the basic tools of MCS. After treating the basic tools of MCS, Chapter 4 examines the crucial elements of analyzing the properties of asymptotic test procedures by MCS. Chapter 5 examines more general aspects of MCS, such as its history, possibilities to increase its efficiency and effectiveness, and whether synthetic random exogenous variables should be kept fixed over all the experiments or be treated as genuinely random and thus redrawn every replication. The simulation techniques that we discuss in the first five chapters are often addressed as naive or classic Monte Carlo methods. However, simulation can also be used not just for assessing the qualities of inference techniques, but also directly for obtaining inference in practice from empirical data. Various advanced inference techniques have been developed which incorporate simulation techniques. An early example of this is Monte Carlo testing, which corresponds to the parametric bootstrap technique. Chapter 6 highlights such techniques and presents a few examples of (semi-)parametric bootstrap techniques. This chapter also demonstrates that the bootstrap is not an alternative to MCS but just another practical inference technique, which uses simulation to produce econometric inference. Each chapter includes exercises allowing the reader to immerse in performing and interpreting MCS studies. The material has been used extensively in courses for undergraduate and graduate students. The various chapters all contain illustrations which throw light on what uses can be made from MCS to discover the finite sample properties of a broad range of alternative econometric methods with a focus on the rather basic models and techniques.

The generalized method of moments (GMM) estimation has emerged as providing a ready to use, flexible tool of application to a large number of econometric and economic models by relying on mild, plausible assumptions. The principal objective of this volume is to offer a complete presentation of the theory of GMM estimation as well as insights into the use of these methods in empirical studies. It is also designed to serve as a unified framework for teaching estimation theory in econometrics. Contributors to the volume include well-known authorities in the field based in North America, the UK/Europe, and Australia. The work is likely to become a standard reference for graduate students and professionals in economics, statistics, financial modeling, and applied mathematics.

Table of contents

This highly accessible and innovative text with supporting web site uses Excel (R) to teach the core concepts of econometrics without advanced mathematics. It enables students to use Monte Carlo simulations in order to understand the data generating process and sampling distribution. Intelligent repetition of concrete examples effectively conveys the properties of the ordinary least squares (OLS) estimator and the nature of heteroskedasticity and autocorrelation. Coverage includes omitted variables, binary response models, basic time series, and simultaneous equations. The authors teach students how to construct their own real-world data sets drawn from the internet, which they can analyze with Excel (R) or with other econometric software. The accompanying web site with text support can be found at www.wabash.edu/econometrics.

Econometric Theory and Methods International Edition provides a unified treatment of modern econometric theory and practical econometric methods. The geometrical approach to least squares is emphasized, as is the method of moments, which is used to motivate a wide variety of estimators and tests. Simulation methods, including the bootstrap, are introduced early and used extensively. The book deals with a large number of modern topics. In addition to bootstrap and Monte Carlo tests, these include sandwich covariance matrix estimators, artificial regressions, estimating functions and the generalized method of moments, indirect inference, and kernel estimation. Every chapter incorporates numerous exercises, some theoretical, some empirical, and many involving simulation.

This book contains an up-to-date coverage of the last twenty years advances in Bayesian inference in econometrics, with an emphasis on dynamic models. It shows how to treat Bayesian inference in non linear models, by integrating the useful developments of numerical integration techniques based on simulations (such as Markov Chain Monte Carlo methods), and the long available analytical results of Bayesian inference for linear regression models. It thus covers a broad range of rather recent models for economic time series, such as non linear models, autoregressive conditional heteroskedastic regressions, and cointegrated vector autoregressive models. It contains also an extensive chapter on unit root inference from the Bayesian viewpoint. Several examples illustrate the methods.