Agent based information aggregation markets

Abstract

Information Aggregation Markets (IAMs) constitute a mechanism whose purpose is to collect and aggregate information. They are commonly known as 'prediction markets' because they are often used to predict future events. In order for IAMs to efficiently aggregate information, they should attract participants with diversity of opinion, independence of thought and decentralization of knowledge. When participants arc typically well-informed, IAM prices will aggregate information into market prices. IAMs can be considered as a large-scale, open, distributed system used by many human participants whose information needs to be aggregated. In this paper we propose a different approach for information aggregation using IAMs. We use autonomous, interacting agents instead of human participants and we show that agent based IAMs can act as an information aggregation mechanism able to perform predictions without human involvement. Copyright © 2010, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.