Discrete Choice Methods with Simulation

Discrete Choice Methods with Simulation by Kenneth Train has been available in the second edition since 2009. The book is published by Cambridge University Press and is also available for download on the author’s homepage for private use. The second edition corrects some errors and contains two additional chapters, one on endogenous regressors and one on the expectation–maximization (EM) algorithm. As the title suggests, the book has two main topics. One concerns models of discrete (multinomial) choices that deal with the decision between a finite set of mutually exclusive alternatives such as the choice between brands of consumer goods, travel modes, or college majors. These models are important for fields such as marketing, transportation, environmental economics, education, labor, and industrial organization. The second major subject of the book is a simulation-based estimation with a focus on maximum simulated likelihood. These two topics are very compatible since multinomial choice models quite naturally lead to likelihood functions and other objective functions that are analytically infeasible to evaluate, requiring approximation methods such as a Monte Carlo simulation. As a consequence, much of the methodological work on simulation-based estimation was inspired by discrete choice models and much of the more recent empirical work studying discrete choices uses simulation-based estimation.