BRA: An Algorithm for Simulating Bounded Rational Agents

Authored by Stephan Schuster

Date Published: 2012-01

DOI: 10.1007/s10614-010-9231-1

Sponsors: No sponsors listed

Platforms: No platforms listed

Model Documentation: Other Narrative Pseudocode Mathematical description

Model Code URLs: Model code not found

Abstract

This paper describes a simulation approach for modelling decision-making processes under incomplete and imperfect information in Agent-based Computational Economics (ACE). The main idea is to represent decision-making in a model-free framework that can be applied to a larger set of simulation problems, not just the domain modelled. The method translates some basic sociopsychological concepts from the bounded rationality and learning literature into an executable algorithm. In a simple example, the algorithm is applied in the domain of behavioural game theory, illustrating how the algorithm can be used to reproduce observed patterns of human behaviour.

Tags

Bounded rationality Agent based modelling reinforcement learning