SÉMINAIRE DU FONDS CONRAD-LEBLANC : Big Data et Deep Learning en Finance

Ruey S. Tsay, Ph.D. (Wisconsin)
H.G.B. Alexander Professor of Econometrics and Statistics
The University of Chicago Booth School of Business

“Ruey Tsay studies business and economic forecasting, big data analysis, risk modeling and management, credit ratings, and process control. Tsay’s research aims at finding the dynamic relationships between variables and how to extract information from messy data. He has authored several best-selling books comme Analysis of Financial Time Series, 3rd Edition, published in 2010 by Wiley; An Introduction to Analysis of Financial Data with R, published in 2012 by Wiley; Multivariate time series analysis with R and Financial Applications, published in 2014 by Wiley; and coauthored A Course in Time Series Analysis, with D. Pena and G. Tiao, published by Wiley in 2001. Tsay has worked as a consultant for numerous American, Chinese, and Taiwanese companies. This experience taught him what works in practice and what does not – knowledge that he shares with students in the classroom. He hopes they learn ideas and methods for extracting information from data, large or small.

Tsay is the winner of the 2005 IBM Faculty Research Award and the John Wiley and Sons Author of the Year for his book, Analysis of Financial Time Series, in probability and statistics. He has received nine National Science Foundation grants and holds a U.S. patent for a system and method for building a time series model. He has delivered invited lectures at IMF and central banks of several countries.

He is a fellow of the American Statistical Association, the Institute of Mathematical Statistics, the Royal Statistical Society, and Academia Sinica. He is a coeditor of the Statistica Sinica, and associate editor of Asia-Pacific Financial Markets, Studies in Nonlinear Dynamics and Econometrics, and Metron.”

AI, Big Data, Statistics, and the Future

Artificial Intelligence (AI, or Deep Learning) and big data have attracted great attention in recent years. The availability of big data and advancements in computation methods and capability further accelerate the development in machine learning. There is no doubt that AI will affect every aspect of our life in the future. In this talk, we discuss the following issues: (a) What is AI? (b) What is machine learning? (c) What roles can data play in the development of AI? (d) How important is statistics in Deep Learning? And (e) How to prepare for the AI challenges? The talk will emphasize on the statistical view on the value of data and the importance of statistical reasoning and methods in the development of smart AI.

Jianqing Fan, Ph.D. (Berkeley)
Frederick L. Moore ’18 Professor of Finance, Professor of Statistics, and
Professor of Operations Research and Financial Engineering Princeton
University

“Jianqing Fan is a statistician, financial econometrician, and data scientist. He is the winner of The 2000 COPSS Presidents’ Award, Morningside Gold Medal for Applied Mathematics (2007), Guggenheim Fellow (2009), Pao-Lu Hsu Prize (2013) and Guy Medal in Silver (2014). He got elected to Academician from Academia Sinica in 2012. He is also the dean of the School of Data Science at Fudan University since 2015.”

“The COPSS Presidents’ Award, sometimes referred to as the « Nobel Prize of Statistics », is given annually by the Committee of Presidents of Statistical Societies to a person under the age of 41, in recognition of outstanding contributions to the profession of statistics.”

“The Royal Statistical Society Guy Medal in Silver is named after the distinguished statistician, William Guy FRS.” Among them are the legendary winners M. S. Bartlett, L. H. C. Tippett, George Box, C. R. Rao.

“Fan is interested in statistical theory and methods in data science, statistical machine learning , finance, economics, computational biology, biostatistics with particular skills on high-dimensional statistics, nonparametric modeling, longitudinal and functional data analysis, nonlinear time series, wavelets , among others. Fan has been consistently ranked as a top 10 highly-cited mathematical scientist.”

“Jianqing Fan is a joint editor of Journal of Business and Economics Statistics and an associate editor of Management Science (2018–), among others, was the co-editor(-in-chief) of the Annals of Statistics (2004-2006) and an editor of Probability Theory and Related Fields (2003-2005), Econometrical Journal (2007-2012), Journal of Econometrics (2012-2018; managing editor 2014-18), and on the editorial boards of a number of other journals, including Journal of the American Statistical Association (1996-2017), Econometrica (2010-2013), Annals of Statistics (1998-2003), Statistica Sinica (1996-2002), and Journal of Financial Econometrics (2009-2012). He was the past president of the Institute of Mathematical Statistics (2006-2009), and past president of the International Chinese Statistical Association (2008-2010).”

Structural Deep Learning in Financial Asset Pricing

We first gives an overview on the genesis of machine learning and artificial intelligence (AI) and how statistical and computational methods have evolved with growing dimensionality and sample sizes and become the foundation of modern machine learning and AI. Then, we develop new financial economics theory guided structural nonparametric methods for estimating conditional asset pricing models using deep neural networks, by employing time-varying conditional information on alphas and betas carried by firm-specific characteristics. Contrary to many applications of neural networks in economics, we can open the “black box” of machine learning predictions by incorporating financial economics theory into the learning, and provide an economic interpretation of the successful predictions obtained from neural networks, by decomposing the neural predictors as risk-related and mispricing components. Our estimation method starts with period-by-period cross-sectional deep learning, followed by local PCAs to capture time-varying features such as latent factors of the model.  We formally establish the asymptotic theory of the structural deep-learning estimators, which apply to both in-sample fit and out-of-sample predictions. We also illustrate the “double-descent-risk” phenomena associated with over-parametrized predictions, which justifies the use of over-fitting machine learning methods. (Joint with Tracy Ke, Yuan Liao, and Andreas Neuhierl).