Multi-agent approaches to economic modeling: Game theory, ensembles, evolution and the stock market

Abstract

A multi-agent system that learns by using neural networks is implemented to simulate the stock market. Each committee of agents, which is regarded as a player in a game, is optimized by continually adapting the architecture of the agents through the use of genetic algorithms. The proposed procedure is implemented to simulate trading of three stocks, namely, the Dow Jones, the NASDAQ and the S&P 500.