Title: Determinants of market volatility: A latent threshold dynamic model
Authors: John Maheu - McMaster University (Canada)
Azam Shamsi Zamenjani - University of New Brunswick (Canada) [presenting]
Abstract: Measuring, modeling, and forecasting volatility are of great importance in financial applications such as asset pricing, portfolio management, and risk management. We investigate the predictability of stock market volatility by macro-finance variables in a dynamic regression framework using latent thresholding. The latent threshold models allow data-driven shrinkage of regression coefficients, by collapsing them to zero for irrelevant predictor variables and allowing for time-varying nonzero coefficients when supported by the data. This is a parsimonious framework that selects what potential predictor variables should be included in the regressions and when. We discuss the Bayesian model specification of dynamic regressions using latent thresholding. We incorporate a large number of potential predictors and let the data determine the relevant predictors over time. We applied the models to monthly S\&P 500 volatility and nd that using macro-finance variables in volatility forecasts enhances model performance statically and economically, particularly when we allow for dynamic inclusion/exclusion of these variables.