Discrete Choice Modeling Training through Theory and Practice

Introduction

Human life is full of choices to select from on daily basis. Hence, discrete choice models analyses individuals’ choice behavior and solves problems in many fields such as agriculture, economics, accounting, health, engineering, environmental management, urban planning, tourism and transportation among other fields. For example, discrete choice modeling is used in agriculture to inform on the best technology or innovation that is beneficial to farmers. In terms of health, discrete choice modeling informs on the preference of health and healthcare. In market research, discrete choice modeling can guide product positioning, pricing, product concept testing. This 5-days course will equip participants with skills on how to use databases to estimate and test discrete choice models as well as gain hands-on experience in using discrete choice techniques for practical applications.

Duration

5 days

Who should attend?

·           Academia

·           Professionals interested in learning new discrete choice techniques (Masters and PhD students in economics, planning, civil engineering, management, behavioral science, health science, and political science. 

·           Staff in development research

·           Consultants in market research, transportation consulting, planning among others

Course objectives

By the end of training, participants will be able to:

·           Understanding discrete choice models and their applications.

·           Choose between a ranges of different models used for predicting multinomial choices.

·           Identify the advantages and disadvantages of the different econometric models. - use the different models in practice and interpret the outcome.

·           Understanding problems of data collection, model formulation, estimation, testing, and forecasting, as learned through case studies of discrete choice methods.

·           Utilizing Stata or R software to estimate and test discrete choice models from real databases.

COURSE CONTENTS

Module1: Basic statistical terms and concepts

·           Introduction to statistical concepts

·           Descriptive Statistics

·           Inferential statistics

·           Research design

·           Sampling

Theoretical foundations of discrete choice models: Theories of choice

·           Random utility theory

·           Lancaster’s theory of characteristics

·           Neoclassic economics

Module 2: Introduction to behavior modeling

·           Analysis of revealed and stated preferences sampling

·           Learning how to use Stata/R software:

·           Binary choice models

·           Probabilistic choice models

·           Logit model

·           Specification of the Logit/Probit model,

·           Estimation of Logit/Probit parameters, the validation process, and their application.

·           Nested logit

Module 3: Choice with multiple alternatives

·           Multinomial logit model

·           Ordered probit/ordered logit model

·           Specification of the Multinomial Logit/Probit  and ordered logit/ordered logit  models

·           Estimation of Multinomial Logit/Probit and ordered logit/ordered logit parameters, the validation process, and their application.

·           Case studies with real data sets,

·           Nested logit

Module 4: Data management and analysis

·           Data analysis using Stata or R

·           Model applications

·           Case studies on estimation of binary choice model with real data sets,

·           Case studies on estimation of Multinomial Logit/Probit and ordered logit/ordered logit with real data sets,

Module 5:  Interpretation and discussion of results

·           Case study: Interpretation and discussion of results from real data analyzed during training