|dc.description.abstract||In the electronic market place, the auction is known to be economically efficient, allowing players to maximize their own benefits. Through this mechanism, new spot markets are created, which connect buyers and sellers. In this new spot markets, many problem contexts give rise to competitive decision situations in which players must make repeated decisions along with or in response to competitors' decisions. Auction-based electronic marketplaces for freight service procurement are an example of such environments, and provide the motivating application context for the models presented in this paper.
The specific focus is on the decisions of carriers, as bidders for the loads tendered by shippers in spot market situations. This paper is about the learning models used to describe a player's strategy choice behavior using experimental data and explains how that choice arises from the nature of multi-player interactions and their dynamics over multiple bids. Therefore, the principal focus of this paper is how to model a player's dynamic strategy choice behavior under the pressure of competition.
A dynamic strategy choice model structure for two type of cognitive learning process is formulated, with alternative specifications corresponding to different levels of cognition capacity. Furthermore, the dynamic strategy choice model structure for mixed learning is developed, which combines both elements of two different learning processes. The model is intended to describe how a player or agent in a non-cooperative game with no perfect information and bounded rationality chooses a bidding strategy. We propose a general dynamic strategy choice model framework using the dynamic mixed logit model structure and estimate the model parametrically, using two sets of experimental data. The paper also presents econometric issues that arise in estimating such models given a time series of auction bids and outcomes, and formulates error structures appropriate to the highly interactive dynamic nature of competitive auction-based marketplaces.||en_US