| Online auctions is to extend the traditional transactions to the Internet platforms,and by expanding the influence of auctions,to attract more bidders to participate in.However,the openness and anonymity of the Internet has brought a series of fraud problems based on Internet auctions.So how to identify and prevent the occurrence of fraud;and in the case of fraud,the bidder’s bidding choices and changes in income have become an important issue in the study of online auction fraud.This thesis mainly studies the shill bidding in online auctions.Auction are divided into three models: independent-private-value model,common-value model and affiliatedvalue model.The existing literature is mostly based on the independent-private-value model,while the research on the common-value model and the affiliated-value auction model is less;secondly,there are fewer researches on the identification of shill bidding.To solve these problems,based on the assumptions of three models,this thesis establishes the expected return model of shill bidding,and solves the optimal equilibrium bidding strategy of bidders.Based on the actual auction data of e Bay,this thesis presents a way to identify the occurrence of shill bidding behavior.The main research contents are as follows:(1)In view of the strategy of bidders and auctioneers,based on three traditional auction models,consider the British auction method of single-item auctions.First,the bidding strategy of the bidders is solved under the condition that the shill bidding behavior does not occur and there is probability of the shill bidding behavior.and the conclusion is drawn that the shill bidding behavior of the auctioneer will make the expected revenue increase faster,and as the number of bidders increases,the expected revenue of the auctioneer will be closer to the actual value of the auction items faster..Secondly,based on the non-cooperative game model of the auctioneer and the bidder,the decision-making situation and Nash equilibrium between the two parties under different game choices are analyzed quantitatively.Finally,it is found that the incomplete rational attitude based on the bidder leads to the winner’s curse.(2)In view of the identification of shill bidding,this thesis collected auction data related to e Bay website for empirical analysis.The arrival process of the bidders is approximated by the non-homogeneous Poisson process,and the arrival of the bidders is predicted under the inverted exponential model.Secondly,based on this model,a method is proposed to identify the bidding behavior by comparing the deviation between the optimal auction duration and the actual auction duration. |
