Chapter 10 - Evaluating Models: Verification, Calibration, Validation
This folder includes the accompanying resources for the chatper. For full book details, see: http://www.abmgis.org/.
Model evaluation is one of the central challenges associated with agent-based models. A key question that all modellers face is "how well does this model simulate the phenomenon of interest?". While there are no universally accepted methods for evaluating agent-based models, researchers often adopt the same three stage process of verification, calibration and validation. This chapter presents an overview of the methods that are commonly used within each of these stages. The overarching aim of this chapter is to provide the reader with the knowledge to design their own approach to evaluating agent-based models.
Tutorial / Lecture Slides
- BehaviorSpace: Simple example taking NetLogo's Traffic Grid model and adding a BehaviorSpace experimentation to it to explore the parameter space.
- Rainfall for verification: Demonstrates how verification can be carred out.
- WalkThisWay: Provides an example of quantitative calibration using a model reimplimented NetLogo (See Appendix A7 for the original MASON model)
For full details about the book, see: http://www.abmgis.org/.