A0561
Title: Dynamic factor stochastic volatility in mean model
Authors: Daichi Hiraki - University of Tokyo (Japan) [presenting]
Yasuhiro Omori - University of Tokyo (Japan)
Abstract: A stochastic volatility in mean (SVM) model is developed within a dynamic factor framework to capture common movements in macroeconomic variables under time-varying uncertainty. Motivated by theoretical considerations in macro-finance, the model allows conditional volatility to directly affect the conditional mean through a volatility-in-mean component. This feature enables the model to account for time-varying risk premiums that are otherwise difficult to capture in standard factor stochastic volatility models. The model is estimated using Bayesian Markov chain Monte Carlo methods and applied to quarterly U.S. macroeconomic data from the FRED-QD dataset. The empirical results illustrate how the SVM structure can be embedded in a latent factor setting to study macroeconomic dynamics under uncertainty, providing a basis for future forecasting and structural analysis.
