Evolutionary Agent-Based Economics Evolutionary Agent-Based Economics

The work in this thesis has been carried out at the Centrum voor Wiskunde en Informatica (CWI) under the auspices of the research school IPA (Institute for Programming research and Algorithmics). Acknowledgments I am grateful to the many people who have helped and supported me in the past few years. First of all, I thank my supervisors Han La Poutré and Hans Amman for their valuable input and the many helpful discussions. I had a very enjoyable time at the Centre for Mathematics and Computer Science (CWI) in the SEN4 group and I am grateful for this. I am also grateful to Herbert Dawid and Joost Kok for their valuable input as members of my kernel committee. The Santa Fe Institute 2002 Complex Systems Summer School was a very inspiring and fun event and working with Carolina Castaldi and Bill Rand provided both interesting research opportunities and a lot of fun. I also thank Carolina and Ben Vermin for being my 'paranimfen'. The course on Economics of Networks by Sanjeev Goyal at the Tinbergen institute proved to be a great inspiration and starting point for an important part of this thesis. On a personal level, I would like to thank my parents for their invaluable love and support throughout the years. Finally, I thank Tomas for being there and making me laugh every day. In the last few years, scientists working in the fields of economics and social science have abandoned the assumption of perfect rationality. Theories of (economic) behavior and decision making often assume that decision makers have all the capabilities and information needed to make an optimal decision, that is that they exhibit perfect rationality. Furthermore, in such theories it is often assumed that agents (decision makers) are homogeneous (i.e., have similar characteristics and preferences) and that interaction is anonymous and global. Theoretical outcomes predicted by these theories often differ from outcomes observed in the real world. However, less restrictive and more realistic assumptions regarding agent behavior often lead to analytically untractable models. Recent advances in computer science have however made it possible to simulate artificial societies and thus study economic models by running computer simulations. This new field of study is called Agent-based Computational Economics (ACE). ACE combines elements and views from economics, social sciences and computer science to study such artificial societies. The decision makers in these societies are called agents. Also, ACE allows us to …