Learning to Trade in an Unbalanced Market

Abstract:     
We study the evolution of trading strategies in double auctions as the size of the market gets larger. When the number of buyers and sellers is balanced, Fano et al. (2011) show that the choice of the order-clearing rule (simultaneous or asynchronous) steers the emergence of fundamentally different strategic behavior. We extend their work to unbalanced markets, confirming their main result as well as that allocative inefficiency tends to zero. On the other hand, we discover that convergence to the competitive outcome takes place only when the market is large and that the long side of the market is more effective at improving its disadvantaged terms of trade under asynchronous order-clearing.

Number of Pages in PDF File: 15

Keywords: Trading protocols, Market design, Allocative efficiency, Genetic Programming

JEL Classification: G10, D82, C63

working papers series