Discrete choice models are widely used for the analysis of individual choice behavior and can be applied to choice problems in many fields such as economics, engineering, environmental management, urban planning, and transportation. For example, discrete choice modeling is used in marketing research to guide product positioning, pricing, product concept testing, and many other areas of strategic and tactical interest. Recent applications to predict changes in demand and market share include areas such as choice of travel mode, coffee brands, telephone service, soft drinks and other foods, financial services, internet access, and choice of durables such as smartphones, tablets, automobiles, air conditioners, and houses. This program also covers methods for online applications where predictions of individual choice behavior are used as inputs for the online optimization and personalization of advertising, recommendations and promotions.
The methods covered include discrete choice models (logit, nested logit, generalized extreme value, probit, logit mixtures, hybrid choice models), data collection, specification, estimation, statistical testing, forecasting, and application. The covered topics include analysis of revealed and stated preferences data, sampling, simulation-based estimation, discrete panel data, Bayesian estimation, discrete-continuous models, menu choice, and integration of choice models with latent variables models.
The course includes lab sessions where participants are provided with discrete choice software to learn how to use real databases to estimate and test discrete choice models taught in lectures and gain hands-on experience in using new discrete choice techniques for practical applications. By examining actual case studies of discrete choice methods, students will become familiar with problems of model formulation, estimation, testing, and forecasting.
It is highly recommended that you apply for a course at least 6-8 weeks before the start date to guarantee there will be space available. After that date you may be placed on a waitlist. Courses with low enrollment may be cancelled up to 4 weeks before start date if sufficient enrollments are not met. If you are able to access the online application form, then registration for that particular course is still open.
Takeaways from this course include:
- Understanding discrete choice models and their applications.
- Learning to apply new discrete choice techniques.
- Understanding problems of data collection, model formulation, estimation, testing, and forecasting, as learned through case studies of discrete choice methods.
- Utilizing commonly available software to estimate and test discrete choice models from real databases.
Who Should Attend:
This program is intended for academics and professionals interested in learning new discrete choice techniques and how to predict choice and forecast demand. It is particularly suited for academics engaged in research, especially PhD students in economics, planning, civil engineering, management, behavioral science, health science, and political science. This course will also benefit professionals in market research, transportation consulting, planning, and any government agencies.
Participants will gain hands-on experience in applying discrete choice software in real-world case studies. A working knowledge of basic statistical methods is needed.
Scholarships are available for graduate students and faculty registering for this course — see Links & Resources below.
Please note that laptops are required for this course. Participants will need to install Biogeme. Tablets will not be sufficient for the computing activities performed in this course.
The course consists of a series of lectures and lab sessions that develop discrete choice concepts and techniques and demonstrate their applications. The labs offer hands-on experience in applying the material covered in the lectures using discrete choice software and real-world data sets.
Topics Covered Include the Following:
Theories of choice, random utility models, probabilistic choice models, alternative model formulations, statistical estimation procedures appropriate for alternative data sources, currently available computer software, tests of validity, forecasting procedures, and examples of empirical applications.
The following subjects will be addressed during the course:
- Choice Behavior
- Binary Choice Models
- Specification and Estimation of Choice Models
- Discrete Choice and Machine Learning
- Stated Preference Methods
- Multinomial Choice Models: Probit and Logit
- Specification Testing
- Aggregate Forecasting and Microsimulation
- IIA Tests
- Nested Logit Models
- Extreme Value Models
- Sampling and Estimation
- Mixture Models
- Simulation-Based Estimation
- Dynamic Choice Models and Panel Data
- Combining Revealed and Stated Preferences
- Models with Latent Variables
- Choice from a Menu
- Bayesian Estimation
- Online applications
- Endogeneity and Self-Selection
- Choice Behavior and the Measurement of Well-Being
View 2019 Course Schedule (pdf, subject to change)
Class runs 9:30 am - 5:00 pm every day.
Special events include a reception for course participants and faculty on Monday night and a dinner on Thursday evening. All evening activities are included in tuition.
ASSISTANT PROFESSOR OF MARKETING, SIMON FRASER UNIVERSITY, BRITISH COLUMBIA
"Discrete choice analysis is one of the most valuable new tools available to marketers interested in understanding and predicting consumers' choices ... It goes beyond current textbook treatments of discrete choice analysis with discussions of state-of-the-art developments in the area and experimental applications."
SENIOR RESEARCH ENGINEER, NTT TELECOMMUNICATIONS NETWORKS LABORATORIES
"Discrete choice analysis is an effective way to evaluate a new service and forecast future demand. I have applied it to the estimation of user preference ... in order to develop new telecommunication services."
MEMBER OF TECHNICAL STAFF, BELLCORE
"This course is an excellent review and introduction to discrete analysis theory, ... provides opportunities for hands-on applications. Further, the instructors are major contributors to this area. I highly recommend it."
COORDINATOR OF ECONOMIC RESEARCH, PERUVIAN TELECOMMUNICATIONS REGULATORY AGENCY
"It's a must-have course for any economist seeking to learn or refresh their knowledge of discrete choice models."
TRAFFIC AND REVENUE ANALYST, CINTRA US
"Top-notch, concise instruction in lecture. Coverage of topics did not get too detailed and allowed more sub-areas to be uncovered. Course materials were excellent and it was good to come back with many days of notes and a textbook to serve as a resource."
GRADUATE STUDENT, TEXAS A&M UNIVERSITY
"Ben-Akiva covered a wide range of topics in discrete choice analysis from basic to recent development within a week. Even though the one week is very short to cover all the subjects, his lecture was very intuitively clear and deep. It was a very good experience, and I hope to take another class at MIT if possible."
RESEARCH ASSISTANT, SUFG
"The course is one of a kind!"
Moshe Ben-Akiva is the Edmund K. Turner Professor of Civil and Environmental Engineering at the Massachusetts Institute of Technology (MIT), and Director of the MIT Intelligent Transportation Systems (ITS) Lab, and Principal Investigator at the Singapore-MIT Alliance for Research and Technology. He holds a PhD degree in Transportation Systems from MIT and was awarded honorary degrees from the University of the Aegean, the Université Lumiére Lyon, the KTH Royal Institute of Technology, and the University of Antwerp. His awards include the Robert Herman Lifetime Achievement Award in Transportation Science from the Institute for Operations Research and the Management Sciences, the Lifetime Achievement Award of the International Association for Travel Behavior Research, the Jules Dupuit prize from the World Conference on Transport Research Society, and the Institute of Electrical and Electronics Engineers ITS Society Outstanding Application Award for DynaMIT, a system for dynamic network management. Ben-Akiva has co-authored two books, including the textbook Discrete Choice Analysis, published by MIT Press, and nearly 400 papers in refereed journals or refereed conferences. He has worked as a consultant in industries such as transportation, energy, telecommunications, financial services and marketing for a number of private and public organizations, including Hague Consulting Group, RAND Europe, and Cambridge Systematics, where he was previously a Senior Principal and member of the Board of Directors. He also was an advisor to Memetrics and ChoiceStream, provided litigation support to Analysis Group and Brattle Group and is the Chief Scientific Advisor to Mobile Market Monitor. He was recently a member of the Future Interstate Highway System Committee of the National Academies of Sciences, Engineering, and Medicine.
This course takes place on the MIT campus in Cambridge, Massachusetts. We can also offer this course for groups of employees at your location. Please complete the Custom Programs request form for further details.
|Fundamentals: Core concepts, understandings, and tools (35%)||35|
|Latest Developments: Recent advances and future trends (40%)||40|
|Industry Applications: Linking theory and real-world (25%)||25|
|Lecture: Delivery of material in a lecture format (80%)||80|
|Labs: Demonstrations, experiments, simulations (20%)||20|
|Introductory: Appropriate for a general audience (20%)||20|
|Specialized: Assumes experience in practice area or field (50%)||50|
|Advanced: In-depth explorations at the graduate level (30%)||30|