Sentiment-driven mean reversion in the 4/2 stochastic volatility model with jumps

With the availability of social networks, specialized forums, and online news, sentiment analysis has become a common and useful technique for the analysis of economic and financial scenarios. Several data-providers have also started computing proprietary sentiment indexes on financial assets to be delivered together with market price and trading volume. We develop a modified version of the mean-reverting 4/2 stochastic volatility model introduced in Escobar-Anel & Gong (2020) to describe the dynamics of commodities. In our specification, jumps are allowed in the asset price dynamics, and the drift coefficient may also switch between regimes related to a sentiment indicator. In this framework, we discuss the distributional characteristics of asset returns, provide a numerical procedure for model estimation, and give some preliminary results on the pricing of European-style derivatives. Finally, the model is fitted to the market data for Gold and Crude Oil.