A Multi-attribute bidding strategy for a single-attribute auction marketplace

• An approach to facilitate multi-attribute bidding in single-attribute auctions is proposed.
• A utility concession curve is applied to assess bid acceptability through time.
• Set of hypotheses for testing the benefits of multi-attribute bidding is stated.
• Hypotheses are tested using data from eBay.
• Relationship between concession curve and auction listing volume is established.

Auctions and negotiations are important mechanisms of economic exchange. Negotiations are less structured as they involve exchange of offers by both parties. They may involve several issues and new issues can be added in the course of negotiations. Auctions, on the other hand are highly structured, and thus they are much amenable to automation. However, most existing auctions involve bidding on a single issue (price). This is an obvious limitation as participants may have different preferences over multiple attributes of a given type of product or service. This paper proposes an approach by which a software agent solution could effectively implement a multi-attribute bidding strategy for single-attribute auction marketplaces. This allows to significantly expand the set of potential auctions to bid on, while assuring the minimum utility level for the bidder at a given point in time. The bidding model is based upon concession-making curve introduced in prior work on electronic negotiations. Using data collected from eBay the paper shows that bidding across several attributes would result in higher utility outcome, and faster results than bidding within a single attribute set.