Read the News, Not the Books : Forecasting Firms'Long-Term Financial Performance

Information Systems and Operations Management

Speaker : Zhu (Drew) ZHANG

Associate Professor of Information Systems   

Ivy College of Business, Iowa State University  

HEC Campus - Jouy-En-Josas - Bâtiment T - Salle 20

Abstract: 

Can one forecast a firm’s future financial performance such as profitability without reading its accounting books? In this talk, we show that mining firm-related events in public news can effectively predict various firm financial ratios. By exploiting state-of-the-art neural architectures, our news-powered deep learning models outperform standard econometric models operating on precise historical accounting data. We also observe forecasting quality improvement in multi-task learning settings, i.e., when multiple financial ratios are predicted simultaneously. Our forecasting models (and byproducts such as attention maps and firm embeddings) benefit various stakeholders with not only quality predictions but also explainable insights, in particular when accounting data is not available.

Bio: 

Zhu (Drew) Zhang is Kingland Faculty Fellow in Business Analytics and Associate Professor of Information Systems at Ivy College of Business, Iowa State University. He received his PhD in Computer and Information Science from the University of Michigan. His expertise includes big data analytics, data/text/web mining, and business intelligence. His research has been published in Management Information Systems Quarterly, Journal of Management Information Systems, ACM Transactions on Management Information Systems, Journal of the American Society for Information Science and Technology, IEEE Transactions and Data and Knowledge and Engineering, IEEE Intelligent Systems, and other leading conferences in artificial intelligence and data mining. Dr. Zhang is a member of the AIS, AAAI, ACL, and ACM and serves on the editorial boards of Journal of Database Management and International Journal of Business Intelligence Research.