Agent based modeling

Complex Societies are the result of the actions and interaction of many individuals. In this class we will explain regularities on the society level, by building computer models of individuals and individual interactions. This can be  best illustrated with a simple example: an ant hive. While the individual ant makes very simple decisions the whole hive creates an incredibly complex society. In social science the behavior is maybe not as simple, but the outcomes of individual behavior are often surprising. The whole is more than the sum of it’s parts. For example we will be recreating the Schelling Segregation Model in class. It explains the racially segregated urban patterns of North America. For this model we will create 10000 artificial people in an artificial neighborhood. These ‘agents’ will be mildly racist, if they are surrounded by more than x% of people of the other race, they move. What we will find is that even very low racism leads to strongly segregated neighborhoods. In general we will answer questions of what the macroscopic outcome of micro decisions are. How does individual behavior lead to an emergent property of the whole society. Agent-Based modeling, a relatively new technique, is frequently used in many fields such as geography, ecology, economics as well as social and political science. Therefore the course will not be theoretical, but focused on enabling the students to program agent-based models.

Target audience: Students who are interested in modeling society or the economy.

Requirements: Programming experience is recommended by not necessary. For those whe haven’t programmed, we recommend to study the python tutorial in advance: https://docs.python.org/3/tutorial/

Topics covered in class: Introduction to Python; Classes/Agents in Python; A Spatial Voting Model; A Model of Racism and Segregation; How to create an ABM science project.