Chapter 5 - Towards empirical ABMs of complex systems#
Video overview
This short video introduces the evidence-focused questions explored in Chapter 5.
The earlier chapters showed how to build agent-based models; this chapter asks how to ground them in real-world evidence. That grounding is what separates a model that merely runs from one credible enough to inform decisions about a real system.
Learning objectives
By the end of this chapter, you will be able to:
Explain the concept of empirical agent-based models (Section 5.1).
Identify the different dimensions along which an ABM can be grounded: the spatial environment, socio-economic-demographic attributes, behavioural attributes, rules of actions and interactions, and social networks (Section 5.2).
Distinguish the types of empirical data source, such as surveys, and determine their best use (Section 5.3).
Apply good practices for empirical ABMs, including model design, validation, and documentation (Section 5.4).
This chapter moves from how to build an agent-based model to how to ground one in real-world evidence. The through-line is a single question: what does it take to make an ABM credible enough to inform decisions about a real system? Answering it means knowing what can be grounded, what data can ground it, and what practices keep the result trustworthy. Work through the four sections in order, beginning with 5.1:
What empirical ABMs are, and why data-driven modelling has become the standard for socio-technical and socio-environmental systems.
The dimensions along which a model can be grounded: spatial environment, socio-economic-demographic and behavioural attributes, rules of interaction, and social networks.
The empirical data sources that feed an ABM, from GIS and census data to surveys, experiments, and large language models, and what each is best used for.
How to build incrementally, calibrate and validate at multiple levels, and document a model so it stays transparent and reusable.
Begin with Section 5.1, Introduction.