A1302
Title: Graph neural networks for forecasting multivariate realized volatility with spillover effects
Authors: Chao Zhang - University of Oxford (United Kingdom) [presenting]
Abstract: A novel methodology is presented for modelling and forecasting multivariate realized volatilities using customized graph neural networks to incorporate spillover effects across stocks. The proposed model offers the benefits of incorporating spillover effects from multi-hop neighbors, capturing nonlinear relationships, and flexible training with different loss functions. Empirical findings provide compelling evidence that incorporating spillover effects from multi-hop neighbors alone does not yield a clear advantage in terms of predictive accuracy. However, modelling nonlinear spillover effects enhances the forecasting accuracy of realized volatilities, particularly for short-term horizons of up to one week. Moreover, results consistently indicate that training with the Quasi-likelihood loss leads to substantial improvements in model performance compared to the commonly used mean squared error. A comprehensive series of empirical evaluations in alternative settings confirm the robustness of the results.
