A1653
Title: A GARCH model with two volatility components and two stochastic factors
Authors: Luca Vincenzo Ballestra - Alma Mater Studiorum University of Bologna (Italy) [presenting]
Enzo DInnocenzo - Alma Mater Studiorum University of Bologna (Italy)
Christian Tezza - University of Bologna (Italy)
Abstract: Several studies provide support for stochastic volatility specifications that incorporate two sources of uncertainty, emphasizing that a simple one-factor model inadequately captures the dynamic dependencies in financial data. A GARCH model with two distinct components is proposed, each driven by an independent, unobservable innovation factor. By introducing a second innovation factor, the model's ability to represent market dynamics is enhanced, resulting in a more accurate description of the variance process and the implied volatility surface. The model improves the two-factor GARCH model recently proposed by allowing spillovers between the volatility components rather than treating them as independent, and sufficient conditions are established for stationarity and ergodicity. Results demonstrate that the approach outperforms existing models in capturing S\&P 500 index returns, both in-sample and out-of-sample. Additionally, it shows comparable performance to the prior study's model when pricing S\&P500 options.
