同质物品网上拍卖的最优设计
发布时间：2018-10-10 11:33
【摘要】：随着电子商务的飞速发展,网上拍卖变得日益流行。在网上拍卖的实践中,拍卖机制的好坏将直接影响网上拍卖物品的有效分配,进而影响拍卖商和拍卖网站的收益,最终影响顾客的购买行为。因此,对网上拍卖机制的研究具有重要的理论和实践意义。 论文结合网上拍卖时间上的异步性和空间上的分布性等特点,在建立网上最优拍卖一般优化模型的基础上,从单阶段网上拍卖最优保留价设计、多阶段网上拍卖最优保留价的设计、多物品单时期网上拍卖最优设计以及混合机制下(并行使用拍卖和固定价格机制)的最优设计四个方面对网上拍卖问题进行研究。 首先,引入网上拍卖的刊登费、佣金、罚金以及投标者随机到达等因素,构建网上拍卖公开保留价和秘密保留价下的拍卖商期望收益最大化决策模型,研究网上拍卖最优公开保留价和最优秘密保留价,并给出最优保留价的一般形式。在此基础上,对最优保留价进行分析,所得结论表明:拍卖商的最优公开保留价与最优秘密保留价相等,拍卖商的最大期望收益也相等。 其次,在单阶段网上拍卖模型的基础上对多阶段网上拍卖的最优保留价进行了研究,给出了多阶段网上拍卖最优保留价序列的一般形式。研究表明,每个相同阶段的最优公开保留价和最优秘密保留价相等,并且在两种不同的保留价下拍卖商在相同的阶段所获得的最大期望收益也相等:当拍卖商进行多阶段拍卖时,随着所在阶段数的增加,拍卖商在该阶段所获得的期望收益单调递减;如果投标者估价服从均匀分布,那么拍卖商在每阶段的最优保留价是单调递减的。 再次,引入投标者随机到达特点以及物品持有成本、广告成本和监控成本,在单时期拍卖中重新构建拍卖商的收益函数,分别在公开保留价和秘密保留价下对保留价、拍卖数量、拍卖时间以及单位广告成本进行了最优设计,并对两种情形下的研究结论进行了比较分析。研究发现,当投标人数大于拍卖物品数量时,拍卖商公开保留价时的期望拍卖结束价格小于秘密保留价时的期望拍卖结束价格,并且当拍卖商设置的保留价在最低水平时,秘密保留价下的期望收益大于公开保留价下的期望收益。 最后,通过构造延期成本函数,对混合机制下的基于阈值战略的顾客决策模型进行扩展,给出一般形式的基于阈值战略的顾客决策模型。研究表明:无论延期成本为线性形式还是指数形式,只要高估价顾客的延期成本函数为拍卖剩余时间的连续的严格增函数,那么高估价顾客到达网站时都会根据阈值战略进行决策。在此基础上,以拍卖持续时间、拍卖物品数量以及固定价格为决策变量,建立基于阈值战略的混合机制下的拍卖商最优决策模型,通过数值分析,提出了混合机制下的最优拍卖设计原则和设计策略,并通过算例对所提出的设计原则和策略进行验证。 论文最后对所做工作进行了总结并对有待进一步研究的问题进行了探讨。
[Abstract]:With the rapid development of electronic commerce, online auctions become more and more popular. In the practice of online auction, the quality of auction mechanism will directly affect the effective allocation of online auction items, thus affecting the income of auctioneer and auction website, and finally affecting the purchase behavior of customers. Therefore, the research of online auction mechanism has important theoretical and practical significance. On the basis of establishing the general optimization model of online auction on the basis of establishing the optimal auction general optimization model for online auction, the optimal reserved price of online auction at multi-stage is obtained based on the general optimization model of online auction. On the basis of four aspects of design, multi-item single-period online auction system and hybrid mechanism (concurrent use of auction and fixed price mechanism), the online auction problem is carried out Firstly, by introducing the factors such as the publishing fee, commission, fine and the random arrival of the bidder, the paper constructs an auction business expectation benefit maximization decision model under the open reserve price and the secret reserve price of the online auction, and studies the optimal open reservation price and the optimal public reserve price of the online auction. Keep the price secret, and give the optimal reserve price. On the basis of the analysis of the optimal reservation price, the conclusion shows that the optimal public reserve price of the auctioneer is equal to the optimal secret reservation price, and the maximum expectation of the auctioneer Secondly, on the basis of single-stage online auction model, the optimal reservation price of multi-stage online auction is studied, and the optimal reservation for multi-stage online auction is given. The general form of the price sequence. The study shows that the optimal open retention price and the optimal secret retention price for each of the same stages are equal, and the maximum expected revenue obtained by the auctioneer at the same stage at two different reserved prices is also equal: as the auctioneer performs a multi-stage auction, as the auctioneer performs the multi-stage auction, Where the number of bidders increases, the expected income obtained by the auctioneer in this stage is monotonically decreasing; if the bidder's valuation is subject to a uniform distribution, the auctioneer's optimal insurance at each stage The retention price is monotonically decreasing. Once again, it introduces the stochastic arrival characteristics of the tenderer and the cost of the goods holding, the advertising cost and the monitoring cost, rebuilds the income function of the auctioneer in the auction of the single period, Under the price, the reservation price, the number of the auction, the auction time and the cost of the unit advertisement are analyzed, and in both cases, The research results show that when the number of bidders is larger than the number of auction items, the auction end price at the time of the auction ending price is less than the expected auction ending price at the time of the secret reservation price is disclosed by the auctioneer, and when the auctioneer is set up, Where the reserved price is at the lowest level, the expected benefit under the secret reserve price is greater than Finally, by constructing the extended cost function, the customer decision-making model based on threshold strategy under the hybrid mechanism is extended, and the general form is given. Based on the threshold strategy, the customer decision-making model is based on the threshold strategy. The research shows that whether the delay cost is a linear form or an exponential form, as long as the extension cost function of the high-valued customer is a continuous strictly increasing function for the remaining time of the auction, the high-valued customer arrives at the network. On this basis, based on the auction duration, the number of the auction items and the fixed price as the decision variables, the optimal decision model of the auctioneer under the hybrid mechanism based on the threshold strategy is established, and the hybrid mechanism is put forward by numerical analysis. The optimal auction design principle and design strategy of the optimal auction are calculated by the calculation example. The design principles and strategies are verified. Finally, the paper summarizes the work done.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2009
【分类号】：F224;F713.359
本文编号：2261612
[Abstract]:With the rapid development of electronic commerce, online auctions become more and more popular. In the practice of online auction, the quality of auction mechanism will directly affect the effective allocation of online auction items, thus affecting the income of auctioneer and auction website, and finally affecting the purchase behavior of customers. Therefore, the research of online auction mechanism has important theoretical and practical significance. On the basis of establishing the general optimization model of online auction on the basis of establishing the optimal auction general optimization model for online auction, the optimal reserved price of online auction at multi-stage is obtained based on the general optimization model of online auction. On the basis of four aspects of design, multi-item single-period online auction system and hybrid mechanism (concurrent use of auction and fixed price mechanism), the online auction problem is carried out Firstly, by introducing the factors such as the publishing fee, commission, fine and the random arrival of the bidder, the paper constructs an auction business expectation benefit maximization decision model under the open reserve price and the secret reserve price of the online auction, and studies the optimal open reservation price and the optimal public reserve price of the online auction. Keep the price secret, and give the optimal reserve price. On the basis of the analysis of the optimal reservation price, the conclusion shows that the optimal public reserve price of the auctioneer is equal to the optimal secret reservation price, and the maximum expectation of the auctioneer Secondly, on the basis of single-stage online auction model, the optimal reservation price of multi-stage online auction is studied, and the optimal reservation for multi-stage online auction is given. The general form of the price sequence. The study shows that the optimal open retention price and the optimal secret retention price for each of the same stages are equal, and the maximum expected revenue obtained by the auctioneer at the same stage at two different reserved prices is also equal: as the auctioneer performs a multi-stage auction, as the auctioneer performs the multi-stage auction, Where the number of bidders increases, the expected income obtained by the auctioneer in this stage is monotonically decreasing; if the bidder's valuation is subject to a uniform distribution, the auctioneer's optimal insurance at each stage The retention price is monotonically decreasing. Once again, it introduces the stochastic arrival characteristics of the tenderer and the cost of the goods holding, the advertising cost and the monitoring cost, rebuilds the income function of the auctioneer in the auction of the single period, Under the price, the reservation price, the number of the auction, the auction time and the cost of the unit advertisement are analyzed, and in both cases, The research results show that when the number of bidders is larger than the number of auction items, the auction end price at the time of the auction ending price is less than the expected auction ending price at the time of the secret reservation price is disclosed by the auctioneer, and when the auctioneer is set up, Where the reserved price is at the lowest level, the expected benefit under the secret reserve price is greater than Finally, by constructing the extended cost function, the customer decision-making model based on threshold strategy under the hybrid mechanism is extended, and the general form is given. Based on the threshold strategy, the customer decision-making model is based on the threshold strategy. The research shows that whether the delay cost is a linear form or an exponential form, as long as the extension cost function of the high-valued customer is a continuous strictly increasing function for the remaining time of the auction, the high-valued customer arrives at the network. On this basis, based on the auction duration, the number of the auction items and the fixed price as the decision variables, the optimal decision model of the auctioneer under the hybrid mechanism based on the threshold strategy is established, and the hybrid mechanism is put forward by numerical analysis. The optimal auction design principle and design strategy of the optimal auction are calculated by the calculation example. The design principles and strategies are verified. Finally, the paper summarizes the work done.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2009
【分类号】：F224;F713.359
【引证文献】
相关硕士学位论文 前1条
1 高宁宁;政府网上采购拍卖的占线问题研究[D];北京化工大学;2012年
，本文编号：2261612
本文链接：https://www.wllwen.com/wenyilunwen/guanggaoshejilunwen/2261612.html
