Competitive Bidding with Dependent Value Estimates

A bidding situation in which there is uncertainty about the value of the item of interest is modeled. The uncertainty is modeled in probabilistic terms, and the model allows the errors of estimation (the differences between expected values and the actual value) of the bidders to be dependent. The effect of this dependence on the “winner's curse” (the tendency for the highest bidder to be one who has overvalued the item) is studied, and optimal bidding strategies are determined.