Dominant strategy double auction mechanisms: Design and implementation.

摘要

Millions of transactions are arranged over online marketplaces every day. Because transactions of identical commodities are carried out in many isolated auctions, a potential customer may need to decide whether to continue competing in the current auction or to give up and move on to the next during the bidding process. To get a good deal, the buyer has to speculate the other bidders' action and the price movement of the item. Since it is not clear how the buyers should bid to maximize their payoffs, this practice places a huge decision burden on the buyers and deters buyer participation. Consequently, the burden involved can outweigh the savings for customers and result in the loss of customers for online marketplaces.; To regain customers, marketplaces have to simplify customers' decision-making. This paper proposes a new multi-stage double auction design approach to accomplish this task, even when shipping and handling costs, and sales taxes are different across various possible transactions. The novel multi-stage approach renders truthful double auction mechanisms, which simplifies customers' decision-making as bidding one's true valuation (private information) is the best strategy for each individual buyer and seller. This design approach is then applied to procurement marketplaces and new truthful double auction mechanisms for procurement auction are proposed. Compared to other known double auction mechanisms, we show that the resulting mechanisms also achieve higher social welfare and individual payoffs.