Agent-based modeling. The Santa Fe Institute artificial stock market model revisited (Q2465105)

The Santa Fe Institute (SFI) artificial stock market was created in a way that it contained many interesting new technologies that had never been tested in economic modeling. The book under review is an important and careful study of some of the issues involved in the workings of the SFI stock market. This book points out in its historical perspective, the SFI market was intended as a computational test bed for a market with boundedly rational learning agents replacing the standard setup of perfectly rational equilibrium modeling common in economics and finance. The SFI market uses two learning mechanisms extensively: the genetic algorithm and classifier system. In several chapters the author explores some of the more important aspects of the SFI classifier implementations. He shows that the SFI classifier is sensitive to certain design characteristics. Under different assumptions about evolution the classifier system behaves very differently from the original SFI model. The author carefully modifies and explores his own operation on mutating trading strategies. Using this modified mutation causes a situation in which the SFI market is much more likely to converge to the rational expectations equilibrium, and the rich technical trading dynamic does not emerge. The results in the original SFI market are clearly sensitive to how mutation is implemented. The book goes on to do a comparative study between mutation operators. A key issue is how many technical trading related rules are evolved, and whether the system is likely to generate lots of technical rules by chance in the evolutionary process. The modified mutation operator does not generate many of these rules, so they never really get a foot in the door of trading activity. The SFI structure facilitates their formation, but it is possible this could be driven more by genetic drift than selection. The original SFI studies never really answered these questions, and it only looked at trading strategy formation in an indirect level by looking at aggregate numbers. This was a clear weakness. The author does some careful checks to see if technical rules are adding value at the agent level. It appears that they are, so many of the SF1 indirect conclusions are sound. Black LeBaron is right if he says: ``This book is an important piece of work for understanding the dynamics of models with interacting learning agents. I think researchers in the future will find it critical in helping them to understand the dynamics of evolutionary learning models. Most importantly, it sets an important standard for doing careful internal experiments on these markets and the learning mechanisms inside them.'' In my opinion, Ehrentreich's book is an excellent reference to both the learning, and empirical literature in finance.