Agent-Based Modelling for Business, Economics, Informatics and Social Science

Abstract:

Agent-based modelling is a methodology with ample applications in data-intensive fields. Its main focus is not about constructing regressive models to fit observed data. Instead it is about understanding the link between micro-level dynamics (local rules of interaction, behaviour) and (emergent) macro-properties. Therefore, it revolts around the conceptualisation and analysis of stylised - and minimalistic - models that capture specific mechanisms at work. 
 
The course covers topics as variegated as: Product adoption, diffusion of opinions, virus diffusion (in social and computer networks), segregation in society, consensus formation (again in social and computer networks), agent behaviour in financial markets. Interestingly, the techniques described are not only valid for the specific systems under consideration, but they can be easily applied to other focal areas of interest. 
 
The course is highly interactive. All the lectures have first a theoretical part, then, the students must develop (in small groups and always supported by the instructors) the models themselves. This allows them to gain direct experience and familiarity with the concepts taught and the techniques involved. In this participatory environment, multiple exercises and the creation of visualisations play an important role

Instructors:

Prof. Dr Claudio J. Tessone (theory and practice)

Dr Manuel S. Mariani (theory and practice)

Jian-Hong Lin (practice)

Type:
Semester course (seminar-like: highly interactive)

Target Audience:
Master students assigned to “Wahlpflichtbereich" BWL 4

Frequency:
Each spring semester

APS (ECTS):
6

Work load statement:

Part Workload Total Time

Course attendance

(Theory)

12 lectures à 2h
(12 sessions)

24h

Course attendance

(Practice)

12 lectures à 3h
(12 sessions)

36h

Home works 3h per session 36h
Literature study Preparation before class 30h
Assignment Preparation and Final Work 54h
Total   180h
 

Content:
Topics typically comprised in the course contents include (but are not limited to): modelling complex systems, theory of emergence, critical behavior, emergence in social systems, cellular automata, game of life, sand-pile model, public-good games, assignment allocation, minority games, voter model, continuous opinion models, game theory signals (policy changes, advertisement)

Language:
English

Prerequisites:
Basic Python and/or R programming skills (or the willingness to develop this knowledge prior to the course) are a necessary requirement. Basic probability theory, linear Algebra

Suggested reading:

Slanina, F. (2013): Essentials of Econophysics Modelling, Oxford University Press.

Grading:
Participation, oral workshop, presentation,group work, screencasts & materials.

Dates and Location:
Vorlesungsverzeichnis

Location: tba

Further information:
Network Science

Market Research

Registration:

Join our courses and make up your mind if you want to participate. Then officially register using the “Buchungstool” at the University of Zurich.

Note:
This information in the syllabus supports the official information in the electronic university calendar (VVZ – “Vorlesungsverzeichnis”). In cases of doubt, the official information at the VVZ is valid.

 

More information is available in the syllabus (PDF, 167 KB)