单物品在线拍卖的价格形成因素研究管理科学与工程专业论文.docx

哈尔滨工业大学管理学硕士学位论文 哈尔滨工业大学管理学硕士学位论文 II II Abstract With the development of online auction market, more and more people start to bid and sell goods at online auction sites. Different with the heterogeneity and uniqueness of goods at traditional auction market, online auction sites provide plenty of homogeneous goods to be bid at the same time. It is interesting to investigate why the identical goods are sold at different prices. Those hidden rules would help us to develop online auction theory , and to better understand the behavior of bidders. Compared with the complete theories in traditional auction market, theories in online auction area are not sufficient. At moment most theories applied to online auctions are based on traditional auction theories, marketing theories and psychology ones, and in some cases traditional theories are refuted in online auction contexts. Thus, empirical studies play a vital part in testing online auction theories. In this paper, we propose a price formation model based on previous studies. We include four factors in our model which are exterior design factor, seller reputation factor, perceptive goods information factor and bidder strategy factor. We then detail our model by choosing the variables for the first three factors and propose several hypotheses based on Taobao context. The bidder strategy factor is the main contribution of our study. To find out its effect on auction price, we build up a theory model and recognized five different bidding types using K-means clustering method. According to the difference of their bidding time, bids number and winning proportion, we propose the hypothesis on how bidder strategies influence auction price. Then we developed two web crawlers using C++ language, to automatically collect data from Taobao real transactions. Afterwards, we test our model and hypothesis using multi-linear regression model, the results prove that our model is highly significant and the price formation process involves a