Title: Uncommon Factors for Bayesian Asset Clusters
Speaker: Lin William Cong
Lin William Cong is the Rudd Family Professor of Management and Associate Professor of Finance at the Johnson Graduate School of Management at Cornell University SC Johnson College of Business. He is also the founding faculty director for the FinTech Initiative at Cornell. Prior to joining Cornell, he was an assistant professor of Finance and Ph.D. advisor at the University of Chicago Booth School of Business and faculty member at the Center for East Asian Studies. He is a a Kauffman Junior Faculty Fellow, a Poets & Quants World Best Business School Professor, a former doctoral fellow at the Stanford Institute for Innovation in Developing Economies, and a former George Shultz Scholar at the Stanford Institute for Economic Policy Research. Cong serves as associate editor for Management Science, Journal of Financial Intermediation, Journal of Corporate Finance, and the Journal of Banking and Finance, has advised FinTech organizations such as Wall Street Blockchain Alliance and ChainLink, was consulted for regulators’ lawsuits against KIN/Kik and Telegram’s TON regarding their ICOs, as well as for the incubation of Dfinity and its initial research. Cong is a member of multiple professional organizations such as the American Economic Association, European Finance Association, and the Econometric Society.
Abstract: Extant model selection methods either assume homogeneous data observations which follow one common model or search in restricted space heterogeneous models for exogenous given subsets of observations. For panel data in economics or finance may require heterogeneous model selection for each (potentially unknown) clusters the observations naturally form. We invent a novel approach to solving the joint problem of observation clustering and model selection.
Our Clustered Bayesian Model (CBM) combines tree-based supervised clustering algorithms and Bayesian modeling with the spike-and-slab prior distributions.First, cross-sectional observations are clustered recursively into leaves by a tree that grows according to the marginal likelihood jointly for all selected leaf models. Second, observations in each leaf fit a model separately with uncommon variables using data in all periods. Third, the Bayesian model allows time-varying coefficients driven by observation subject characteristics under modest computational costs.
We apply CBM to the (imbalanced) panel of individual stock returns for estimating and selecting observable factor models. CBM splits cross-sectional stock returns by firm characteristics and selects potentially distinct factor models for each leaf clusters. Empirically, we find most asset clusters can be explained by the list of published factors, but some have significant alphas. CBM provides a graphical tree-leaf path with firm characteristics to analyze these mispriced stocks. Finally, we provide Bayesian inference on factor usefulness and the fundamental and macroeconomic sources of mispricing clusters.
Time: 9:00- 10:30 am, Friday, Oct 5, 2022
Venue: ZOOM https://us02web.zoom.us/webinar/register/WN_uahP7245SCqT7OkJQ95qWA
