Competition among intelligent agents and standard bidders with different risk behaviors in simulated English auction marketplace

Agents have been used to bid in online auctions to take over the role of human bidders. We can find bidder agents with a variety of bidding strategies that participate in online auction. However, it is not known how the presence of these agents will affect the marketplace in terms of closing price and the chances of winning. In this paper, we study the economic consequence from a simulated English auction market populated by intelligent agents and three groups of standard bidders with different risk preferences. Our study revealed that when intelligent agents compete with the standard bidders, these agents generally perform better than their counterparts. More specifically, we analyse their average winning utility, the average closing price of auctions and the number of auctions won by them. Based on our experimental results, the intelligent agents outperformed the standard bidders in all three aspects.