Please enable JavaScript for this site to run correctly.

Bayesian Agent-Based Population Studies

Model code

Code for the initial version of the model (Routes and Rumours, with a graph-based topology), described in detail in the Hinsch and Bijak (2019) ALife conference paper: current version of the Julia source code, and a static release of the code corresponding to the version presented in the paper.

The current documentation of the model is available from the link below

  Routes and Rumours Model documentation.pdf


Data sources and assessment

Over the course of the project, we have assembled an inventory of metadata regarding various quantitative and qualitative sources of information for use in agent-based migration modelling of a migration test case (Syrian asylum migration to Europe), together with an assessment of their reliability and usefulness for modelling. The searchable online version of the inventory is accompanied by a background paper on assessing the sources of data and knowledge for modelling asylum migration, which available below together with a static version of the inventory, current as of November 2019, retained for completeness.

  Background paper Data and knowledge.pdf

  Metadata inventory.pdf


Cognitive experiments

Protocols and documentation related to cognitive experiments aimed to learn about decision-making processes and the associated biases in the context of migration decisions

  Background paper Decision making.pdf


Mechanisms and design

Description of identified mechanisms generating migration patterns and of the implied design of the agent-based simulation model and simulation experiments

  Background paper Modelling migration routes.pdf


Computational aspects

Formalisms for a bespoke, domain-specific modelling language, integrating agent-based models, the principles of experimental design and model choice, and computer code for the model and its analysis, and documentation of the modelling process by using the provenance approach.

  Background paper Computational methods.pdf