Advanced choice modelling course - July 2017
Full Registration Fee
Full Time Academics
Taught by experts from the University of Leeds, the course will consist of a mixture of lectures, computer practicals (using R), and detailed case studies. Bringing together expertise from fields as diverse as transport, health, marketing and environmental economics, the course will cover all the steps required for successful estimation of flexible Mixed Logit and hybrid choice models, the implementation and interpretation of the Expectations Maximisation algorithm and Bayesian estimation procedures, and an introduction to models of multiple and continuous choice.
After taking this course, participants will be able to estimate and contrast state-of-the-art models, understand the properties and recognise limitations of maximum likelihood estimation and use alternative estimation techniques best suited for their particular research question and dataset. By conducting hands-on exercises with open source software, participants will become familiar with the theories and models, which adds greatly to the learning experience.
Delegates are expected to bring their own laptops. We will have a limited number of laptops available for rental for the duration of the course. Delegates need to book these when registering for the course.
Please book early to avoid disappointment as spaces are limited and the 2016 course sold out quickly.
For further details, please go to http://www.cmc.leeds.ac.uk/courses-phds/cpd/
Fees include three lunches, one evening dinner and access to software and data for the course.
The Centre for Choice Modelling (CMC) at the University of Leeds is a large multi-disciplinary centre bringing together expertise in choice modelling and the study of human decision making from across different fields.
Outline programme for the course
Day 1: Introduction to estimation in R: MNL and basic Mixed Logit - Advanced Mixed Logit topics: distributions, correlations, estimation, WTP, posterior analysis - Advanced Mixed Logit topics in R – Alternative decision rules - Hybrid Choice Models (theory & application issues)
Day 2: Estimation of Hybrid Choice Models in R - Local optima, alternative estimation routines, advanced diagnostics, error calculations - Advanced estimation routines and diagnostics in R
Day 3: Bayesian Estimation - Bayesian estimation in R – Moving beyond discrete choice – MDCEV in R