Discrete Choice Models

This course is focused on learning the properties of discrete choice models, how to estimate them, and their applications. Topics include multinomial logit, nested logit, mixed logit (random coefficient logit), and dynamic discrete choice models. It includes discussions about some standard models in empirical industrial organization and labor economics, such as models in McFadden (1974), Rust (1987), and Keane and Wolpin (1997). Assignments include programming exercises as well as analytical ones.

MIMA Search

授業計画

Introduction to course + discrete choice models Properties of discrete choice models Logit GEV/Nested logit Mixed (random coefficient) logit Numerical maximization Drawing from densities Simulation-assisted estimation EM algorithms Intro to dynamic discrete choice models Rust (1987) Keane and Wolpin (1997)

授業の方法

lectures

成績評価方法

assignments + either referee report or final exam (depending on class size)

教科書

Train, Kenneth E. Discrete choice methods with simulation. Cambridge University Press, 2009.

参考書

McFadden, Daniel. "The measurement of urban travel demand." Journal of public economics 3.4 (1974): 303-328. Keane, Michael P., Petra E. Todd, and Kenneth I. Wolpin. "The structural estimation of behavioral models: Discrete choice dynamic programming methods and applications." Handbook of labor economics. Vol. 4. Elsevier, 2011. 331-461. Rust, John. "Optimal replacement of GMC bus engines: An empirical model of Harold Zurcher." Econometrica: Journal of the Econometric Society (1987): 999-1033. Keane, Michael P., and Kenneth I. Wolpin. "The career decisions of young men." Journal of political Economy 105.3 (1997): 473-522.

履修上の注意

One should have taken a graduate-level introductory econometrics course or its equivalent. Familiarity with basic discrete choice models, such as probit and logit models with binary response, is encouraged.