CMStatistics 2021: Start Registration
View Submission - CFE
A0484
Title: Bayesian network analysis for financial risk management Authors:  Mike So - The Hong Kong University of Science and Technology (Hong Kong) [presenting]
Abstract: A Bayesian network is a probabilistic graphical model that models conditional dependence (causation) among variables. In most cases, the true underlying structure of a set of variables is unknown. The number of possible structures grows explosively when we have more variables in the networks. We apply a new MCMC sampling scheme for structural learning of Bayesian networks to analyze financial returns data. Time-series properties of the Bayesian networks are investigated to understand the risk evolution in financial markets. We illustrate our approach using stock data in Hong Kong.