A Neural Stochastic Volatility Model

TL;DR

This paper introduces a neural stochastic volatility model combining deep recurrent neural networks to improve volatility estimation and prediction in financial time series, outperforming traditional methods.

Contribution

It presents a novel stochastic recurrent neural network framework for modeling volatility dynamics, integrating generative and inference networks for better accuracy.

Findings

Outperforms GARCH and variants in volatility prediction.

Achieves lower negative log-likelihood on stock datasets.

Demonstrates effectiveness of neural stochastic models in finance.

Abstract

In this paper, we show that the recent integration of statistical models with deep recurrent neural networks provides a new way of formulating volatility (the degree of variation of time series) models that have been widely used in time series analysis and prediction in finance. The model comprises a pair of complementary stochastic recurrent neural networks: the generative network models the joint distribution of the stochastic volatility process; the inference network approximates the conditional distribution of the latent variables given the observables. Our focus here is on the formulation of temporal dynamics of volatility over time under a stochastic recurrent neural network framework. Experiments on real-world stock price datasets demonstrate that the proposed model generates a better volatility estimation and prediction that outperforms mainstream methods, e.g., deterministic models such as GARCH and its variants, and stochastic models namely the MCMC-based model \emph{stochvol} as well as the Gaussian process volatility model \emph{GPVol}, on average negative log-likelihood.