【12月2日】【商学院学术论坛】Sales Forecasting by Utilizing Big Data from Both Internal and ExternalSources发布日期：2019-09-12 21:08:10
主讲题目：Sales Forecasting by Utilizing Big Data from Both Internal and ExternalSources
Firms, such as online retailers, typically have plenty information on the visits andtransactions made by their own customers at their own sites. Yet their efforts of applying Big Data technology in sales forecasting is limited by the lack of information from external sources, i.e., firms usually do not know the activities of the customers at other firms. Some third party data collectors have seized the opportunities by providing firms with customer behavior (including visits and purchases) information from external sources. A challenge, however, is that the value of such external data is largely unknown. In this paper, we propose a novel framework and conduct a series of tests to examine the value of customer behavior information from external sources as supplement to firms’ internal data in the context of sales forecasting. The empirical results suggest that higher forecasting accuracy is usually achieved by combining data from both internal sources and external sources than utilizing internal data alone, yet the values of external data vary by their sources and may not increase monotonically with the size of the data.
PhD student from Guanghua School of Peking University. He got the bachelor and master degrees in computer science at Beijing Normal University and Peking University respectively. With strong training in computer science and marketing modeling, his research interests focus on big data and corresponding applications, using traditional marketing model or machine learning approaches. Meanwhile, he is dedicated to some projects, revealing the casual inference of consumer’sresponse to marketing campaigns through field experiments.