From Deep Learning to Deep Econometrics

Calculating true volatility is an essential task for option pricing and risk management. It is made difficult, however, by market microstructure noise. Particle filtering has been proposed to solve this problem because it has favorable statistical properties, but it relies on assumptions about underlying market dynamics. Machine learning methods have also been proposed but lack interpretability and often lag in performance. In this article, the authors implement the stochastic volatility (SV)-PF-RNN: a hybrid neural network and particle filter architecture. Their SV-PF-RNN is designed specifically with stochastic volatility estimation in mind. They demonstrate that it can significantly improve the performance of a basic particle filter.