Skip to content
All library documents

Combining LSTM Networks with Realized GARCH for Volatility Forecasting

Article arXiv papers · Author: Chen Liu et al.
Read the original

Summary

This framework combines realized volatility measures with an LSTM-enhanced realized GARCH model to jointly represent returns and realized volatility. It uses Sequential Monte Carlo for Bayesian inference and forecasting, bringing together financial econometric modeling, high-frequency data, and deep learning. The study evaluates the framework for volatility prediction, tail-risk forecasting, and option pricing, alongside model fit measured with marginal likelihood.

The empirical analysis covers 31 widely traded stock indices over a period that includes the COVID-19 pandemic. The authors report strong in-sample fit and better predictive performance than several benchmark models, as well as adaptation to familiar features of volatility. The description does not identify the benchmarks or provide metric values, so the magnitude and robustness of the reported gains cannot be assessed from this account alone.

Key ideas

  • The model jointly describes returns and realized volatility using an LSTM-enhanced realized GARCH structure.
  • Bayesian inference and forecasting rely on Sequential Monte Carlo.
  • Evaluation spans volatility forecasts, tail risk, option pricing, and marginal likelihood.
  • The empirical study uses 31 stock indices and reports gains against several unspecified benchmarks.

Tags

VolatilityMachine learningStatisticsOptions
Full text
# Deep Learning Enhanced Realized GARCH


# Deep Learning Enhanced Realized GARCH









We propose a new approach to volatility modeling by combining deep learning (LSTM) and realized volatility measures. This LSTM-enhanced realized GARCH framework incorporates and distills modeling advances from financial econometrics, high frequency trading data and deep learning. Bayesian inference via the Sequential Monte Carlo method is employed for statistical inference and forecasting. The new framework can jointly model the returns and realized volatility measures, has an excellent in-sample fit and superior predictive performance compared to several benchmark models, while being able to adapt well to the stylized facts in volatility. The performance of the new framework is tested using a wide range of metrics, from marginal likelihood, volatility forecasting, to tail risk forecasting and option pricing. We report on a comprehensive empirical study using 31 widely traded stock indices over a time period that includes COVID-19 pandemic.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.