报告题目 | Industry and Peer Firm Identification through Usage Mining of XBRL GAAP Taxonomy | ||
报告人（单位） | Hongwei Zhu（University of Massachusetts Lowell） | ||
点评人（单位） | 刘晓星教授 （太阳集团tyc539） | 点评人（单位） | 尹威副教授 （太阳集团tyc539） |
在线会议信息 | 2020年12月1日 下午7：30； 腾讯会议ID：438 940 678；入会密码：201201 | ||
报告人简介 | |||
Hongwei Zhu is a professor of Management Information Systems and the chair of the Operations and Information Systems Department at the University of Massachusetts Lowell. His research focuses on data quality and analytics, with applications in finance and accounting. His work has appeared in MIT Sloan Management Review, Journal of Management Information Systems, and transactions and journals of ACM and IEEE. He is an associate editor of the ACM Journal of Data and Information Quality. | |||
报告内容提要 | |||
Many areas of research and practice require the identification of economically related firms. Known issues of existi ng industry classification systems have engendered a stream of research that attempts to identify industries and peer firms from a variety of data sources. We advance this research by developing a novel method to identify and group related firms based on their selections of standardized tags defined in the XBRL GAAP Taxonomy. We construct a unique form of big data from the tags selected by each firm. Applying an improved spectral coclustering method, we simultaneously identify firm clusters and tag clusters. Our evaluation shows that firm clusters have high within-cluster homogeneity and between-cluster heterogeneity in terms of commonly used business characters. The corresponding tag clusters further reveal common characteristics of the firms. Comparing to rigid and relatively static classification systems, our approach can construct any desired number of clusters that are continuously updated as soon as new financial statements become available. | |||
