Transaction Behavior in Large Networked Financial Databases

-
TBD
  1. Home
  2. Events
  3. Transaction Behavior in Large Networked Financial Databases

Host

Computer Science Department

Speaker Bio: J. Christopher Westland is a Professor in the Department of Information & Decision Sciences at the University of Illinois. He holds a BA in Statistics and an MBA in Accounting. He received his Ph.D. in Computers and Information Systems from the University of Michigan. Dr. Westland has professional experience in the US as a certified public accountant and as a consultant in technology law in the US, Europe, Latin America and Asia. Dr. Westland is the author of numerous academic papers and of seven books: Financial Dynamics (Wiley 2003); Valuing Technology (Wiley 2002); Global Electronic Commerce (MIT Press 2000); Global Innovation Management (Palgrave Macmillan 2008); Red Wired: China’s Internet Revolution (Marshall Cavendish, 2010); Modern Path Analysis & Structural Equation Modeling (Credere 2012); and Financial Auditing with Information Technology (Credere, 2014). Dr. Westland is the Editor-in- Chief of Electronic Commerce Research (Springer), co-Editor of ECRA (Elsevier) and has served on editorial boards of many other information technology journals including Management Science, ISR, IJEC, SSRN. He has served on the faculties at the University of Michigan, University of Southern California, Hong Kong University of Science and Technology, Tsinghua University, University of Science and Technology of China, and other academic institutions. In 2012 Dr. Westland received High-Level Foreign Expert status in China under the 1000-Talents Plan and is currently Overseas Chair Professor at Beihang University. Dr Westland has advised on valuation and technology strategy for Microsoft, Intel, Motorola, V-Tech, Aerospace Corporation, IBM, Pacific Bell, and other technology firms.

Description

Abstract: The distribution of market transaction data is fundamental to decision-making in many important activities in modern corporations. These provide a foundation for valuation in networked markets such as NYSE and NASDAQ; for audit decisions required annually by the SEC; for guidance provided to investors; for investment, marketing strategy, production scale, and other core decisions. Two assertions are posited and tested in this research: (1) that asset prices should be more informative than returns, which are smoothed and transformed analogues of raw prices; (2) that prices are better modeled with mixture Poisson-Gamma distributions than Gaussian distributions; and (3) that mixture Poisson-Gamma distributions provide better predictions of future market volumes and values. These three assertions were supported by the research. Model performance improvement with price versus returns data was substantial: from less than one-quarter of variance, to almost two-thirds. Tweedie-GLM risk assessments were found to be a magnitude smaller than traditional risk assessments, lending support to market inefficiency and increased risk from idiosyncratic factors. The results of this research will become increasingly important with financial datasets growing exponentially ‘bigger’ and require clear roles for models, algorithms, data and narrative which can only be achieved with statistical protocols and algorithmic languages.

Event Topic

Data Science

Tags: