Module Catalogue 2024/25

Pre Requisite Comment

Student NOT on the Transport Stream should have previously completed a Research Method Module.

The course only requirement is some basic knowledge of statistics (t-test, sampling problems etc) and basic econometric knowledge (like linear regressions).

Although not a requisite, it would be useful if students had some knowledge about data collection or empirical experiments at individual level. Students from different master programs can get this background knowledge from different courses. But it is not a pre-requisite.

Co Requisite Comment

N/A

Aims

The aim of the course is to make the student capable of understanding, modelling and forecasting how individuals make decisions.

Understanding how people makes decisions, and what drives their choices is a key questions in several disciplines such as transport, economics, sociology, psychology, marketing, environment.

In particular:
-The course provides students with knowledge on the fundamentals of the decision-making theory with a particular emphasis on understanding the different views of the decision-making process between the two approaches of perfect and bounded rationality, pros and cons of each approach the importance to move toward a unified theory of the decision process.

-The course provides then the knowledge on the mathematical models most commonly used to simulate and forecast individual choices, starting from the very popular discrete choice models (which are based on the neoclassical economic assumption of rational decision makers) and extending them to account for several effects of bounded rationality.

-Students will learn how to specify and estimate these models, how to evaluate which model provides the best representation of the phenomenon under study, how to apply the models to forecast the demand for different policy scenarios and how to compute user benefits for economic valuation.

-A brief RECAP from the module CEG8435 (Data collection and survey methods) will be provided on the type of data needed to understand and model individual choices.

-The theoretical part is supported by an extensive empirical work where students have the possibility to practically estimate the models discussed in the lectures, using real sets of data, and apply them to forecast the demand. The software Python Biogeme or Panda Biogeme is used for this purpose and thought as part of the practical work.

-The course also provides information on how to write a report to present and explain results to policy makers or industrial clients.

Outline Of Syllabus

The themes of lectures delivered are (it does not correspond to the exact lecture schedule):

Introduction:
-Why it is important to study decision process and model individual choices
-Understanding and forecasting individual choices
-Examples from various disciplines (transport, health, economics, marketing, urban etc.).

Theory of individual choice:
-Definition of decision maker, choice set, feasibility of alternatives and constraints
-Concept of utility, compensatory decisions and maximisation
-Non compensatory choice and other heuristics
-The role of behaviour, attitudes, goals and social influence in the decision process
-Behaviour changes and preference formation

Mathematical models:
-Microeconomic and mathematical derivation of the basic discrete choice models (multinomial logit model)
-Modelling preference heterogeneity among individuals, correlation among alternatives and a variety of substitution patterns (the mixed logit model)
-Utility specification and model estimation in practice
-Modelling effects of bounded rationality, such as habit/inertia effect, attitude and perceptions (the hybrid choice model)
-How to choose the best model: when it is appropriate/correct to use each type model
-Modelling diffusion of innovations, such as electric vehicles, automated vehicles, etc.
Understanding and using multiple data sources:
-RECAP of the data: Psychological indicators, Revealed and Stated preference data, Cross sectional, short and long Panel data
-Modelling with multiple data sources

Using the models in practice:
-How to validate the performance of the models estimated
-How to define policy scenarios
-How to apply the models estimated to forecast the demand under different scenarios
-How to use the models estimated for economic valuation: Value of time and user benefits.

Intended Knowledge Outcomes

At the end of the module a student will be able to:

-Understand the basic concept of microeconomic theory and bounded rationality
-Understand the link between the mathematical models and the phenomenon they are able to reproduce, i.e. pros and cons (their strengths and weaknesses) of each model
-Understand the nature of the data that can be used to estimate and forecast individual choices and being able to select the right data for the right phenomenon and/or problem under study.
-Select and use the appropriate mathematical model to study individual choices and forecast the impact of policies, arguing the pros and cons of the models employed.
-Use the software to estimate both standard and advanced DCMs
-Formulate and analyses advanced problems that involved individuals’ choices to identify optimal, sustainable and equitable solutions.
-Interpret and compared the results based on statistical analysis
-Argue concerning the usefulness of the different models for a specific problem
-Read and discuss some papers where some advanced discrete choice models at their choice is estimated
-Work independently as well as in a team

Intended Skill Outcomes

At the end of the module a student will be able to:

-For each specific problem under study (involving individual choice), identify the adequate model structure (among those discussed in the lectures) to simulate the phenomenon and justify the theoretical reasons behind the choice.
-Build (i.e. specify, estimate and evaluate) the mathematical models discussed in the lectures to simulate individual choices in a variety of contexts.
-Using statistical tests and theoretical consideration to evaluate the best model (among those discussed in the lectures).
-Using the models estimated (among those discussed in the lectures) to forecast the demand under various policies.
-Using the models estimated (among those discussed in the lectures) to compute the demand elasticity and valuation measures.
-Identify if a dataset is adequate for the simulation of the phenomenon and/or problems under study
-Write a methodologically sound report containing a description of the phenomenon, treatment of data, a theoretical description of the model specifications (including both standard and advanced DCMs), including discussion on the microeconomic and the identification properties, estimation and a critical discussion of the results.

Teaching Rationale And Relationship

Teaching and learning of this module is done by a combination of lectures, computer demonstrations and practical work, guest lectures, coursework and reading materials. This is in line with the learning outcomes. Lectures, guest lectures and coursework are intended to provide the theoretical background, computer demonstrations allow students to learn the software and the codes to build the mathematical model. Practical work allows students to learn how to link the theory with the practice (how to use the theory in practice) and help developing problem solving skills.
Reading materials helps developing critical, independent and innovative thinking.

Assessment Rationale And Relationship

The two forms of assessment (written exam and report) are intended to serve two purposes:
(1)testing if students acquired the intended skills in terms of understanding and master the theoretical background (written exam) and being able to apply the theory in practice (report);
(2)allowing students, who have different background, to find in one of the three form of examination the form they are more familiar with and where they can express themselves in the best way.

General Notes

N/A