Discrete Choice Modeling
Professor William Greene
Stern School of Business, New York University
National University of Ireland, Galway
with funding from NUI Galway's Millennium Fund
July 4-6, 2012
Prof. William Greene e-mail: firstname.lastname@example.org
Professorís Home Page: http://www.stern.nyu.edu/~wgreene
This course will survey techniques used in modeling discrete data. Discrete choice models can be found in practice in all social sciences, medical research, transport research, the physical sciences, and throughout the research environment where the behavior of individual entities and decision makers is analyzed empirically. This course will present a large number of models and techniques used in these studies. Emphasis will be placed on the most up to date developments in discrete choice modeling.
The presentation, over three days, will include roughly ten morning classroom meetings. In the afternoon of each day, we will do some hands on analysis using "live" data sets and a familiar computer package.
The course will be focused on models and methods. We will do some computation to illustrate some of the model specifications. However, the course is not intended to be a "how to" for using specific techniques with a particular program (e.g., "How to analyze panel data with ____.")
This is a summer school in econometric analysis of discrete choice data. There are a huge variety of models used in this context. The subject matter of the curriculum will focus on four models which arguably comprise the foundation for the area: the fundamental model of binary choice (and a number of variants); models for ordered choice; the Poisson regression model for count data; and the fundamental model for multinomial choice, the multinomial logit model. Many different extensions of the discrete choice models to panel data will be examined. In addition, attention will be paid to a large variety of specifications, particularly those designed to handle heterogeneity, censoring, truncation and selection. Variants on the multinomial logit model will be discussed including recent advances in mixed logit (random parameters), nested logit, and so on, for models that are designed to accommodate a wide variety of behavior patterns.
Discussions will cover the topics listed below.
Prior knowledge is assumed to include calculus and econometrics at the level assumed in the first year of a Ph.D. program in economics and using a textbook such as Greene, W., Econometric Analysis, 6th edition.
No specific textbook is assigned for the course. Useful references are:
Greene, W., Econometric Analysis, 6th Ed., Prentice Hall, 2008 or 7th Ed., 2011
Cameron, A.C. and P. Trivedi, Microeconometrics: Methods and Applications, Cambridge University Press, 2005.
A recently published reference for some of the discrete choice models is
Greene, W. and D. Hensher, Modeling Ordered Choices, Cambridge University Press, 2010.
A lower level textbook that discusses some of the topics we will visit is
Wooldridge, J., Introduction to Econometrics: A Modern Approach, Southwestern, 2008
Part 0: Introduction
Part 1: Linear Regression Models, Discrete Choices, Endogeneity
Part 2: Discrete Choice Models, Binary Choice Models
Part 3: Nonlinear Panel Data Models, Panel Data Binary Choice Models
Part 4: Bivariate Choice Models, Ordered Choice Models, Sample Selection
Part 5: Count Data and Two Part Models
Part 6: Heterogeneity, Mixed Models, Latent Class, Panel Data, Spatial Correlation
Part 7: Multinomial Choice Models
Part 8: Extensions of Multinomial Choice Models, Parameter Heterogeneity
Part 9: Multinomial Choice Models: Willingness to Pay, Stated Choice Experiments
Part 10: Summary
Course material and resources will be available shortly.