Agent-based models of financial markets

Abstract

This review deals with several microscopic (``agent-based'') models of\nfinancial markets which have been studied by economists and physicists over the\nlast decade: Kim-Markowitz, Levy-Levy-Solomon, Cont-Bouchaud, Solomon-Weisbuch,\nLux-Marchesi, Donangelo-Sneppen and Solomon-Levy-Huang. After an overview of\nsimulation approaches in financial economics, we first give a summary of the\nDonangelo-Sneppen model of monetary exchange and compare it with related models\nin economics literature. Our selective review then outlines the main\ningredients of some influential early models of multi-agent dynamics in\nfinancial markets (Kim-Markowitz, Levy-Levy-Solomon). As will be seen, these\ncontributions draw their inspiration from the complex appearance of investors'\ninteractions in real-life markets. Their main aim is to reproduce (and,\nthereby, provide possible explanations) for the spectacular bubbles and crashes\nseen in certain historical episodes, but they lack (like almost all the work\nbefore 1998 or so) a perspective in terms of the universal statistical features\nof financial time series.\n