How Specialised Are Specialists? Generalisation Properties of Entries from the 2008 and 2009 TAC Market Design Competitions

Unlike the classic Trading Agent competition (tac), where participants enter trading strategies into a market, the tac Market Design Competition (cat) allows participants to create rules for their own double auction market and set fees for traders, which they embody in agents known as specialists. Although the generalisation properties of traders when the specialist (i.e., the market mechanism) is fixed have been assessed, generalisation properties of specialists have not. It is unclear whether and how a specialist might (intentionally or unintentionally) favour certain trading strategies. We present an empirical analysis of specialists’ generalisation abilities in various trading environments. Our results show that specialists can be sensitive to a number of factors, including the other trading and specialist strategies in the environment.

[1]  Peter McBurney,et al.  Characterizing effective auction mechanisms: insights from the 2007 TAC market design competition , 2008, AAMAS.

[2]  Nicholas R. Jennings,et al.  IAMwildCAT: The Winning Strategy for the TAC Market Design Competition , 2008, ECAI.

[3]  Michael P. Wellman,et al.  Approximate Strategic Reasoning through Hierarchical Reduction of Large Symmetric Games , 2005, AAAI.

[4]  A. Roth,et al.  Predicting How People Play Games: Reinforcement Learning in Experimental Games with Unique, Mixed Strategy Equilibria , 1998 .

[5]  L. Moses,et al.  Think and Explain with Statistics , 1988 .

[6]  Elizabeth Sklar,et al.  Some preliminary results on competition between markets for automated traders , 2007 .

[7]  D. Cliff.  Minimal-Intelligence Agents for Bargaining Behaviors in Market-Based Environments , 1997 .

[8]  L. Devroye.  Non-Uniform Random Variate Generation , 1986 .

[9]  Robert E. Marks,et al.  Validating Simulation Models: A General Framework and Four Applied Examples , 2007 .

[10]  S. Parsons,et al.  Overview of CAT: A Market Design Competition , 2009 .

[11]  V. Podobnik,et al.  Designing an Effective E-Market : An Overview of the CAT Agent , 2008 .

[12]  Peter Tiño,et al.  Measuring Generalization Performance in Coevolutionary Learning , 2008, IEEE Transactions on Evolutionary Computation.

[13]  John Dickhaut,et al.  Price Formation in Double Auctions , 2001, E-Commerce Agents.

[14]  Peter McBurney,et al.  An Analysis of Entries in the First TAC Market Design Competition , 2008, 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology.

[15]  Dhananjay K. Gode,et al.  Allocative Efficiency of Markets with Zero-Intelligence Traders: Market as a Partial Substitute for Individual Rationality , 1993, Journal of Political Economy.

[16]  H. Robbins.  Some aspects of the sequential design of experiments , 1952 .

[17]  Bart Selman,et al.  A principled study of the design tradeoffs for autonomous trading agents , 2003, AAMAS '03.