Abstract:Volatility, which indicates the dispersion of returns, is a crucial measure of risk and is hence used extensively for pricing and discriminating between different financial investments. As a result, accurate volatility prediction receives extensive attention. The Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model and its succeeding variants are well established models for stock volatility forecasting. More recently, deep learning models have gained popularity in volatility prediction as they demonstrated promising accuracy in certain time series prediction tasks. Inspired by Physics-Informed Neural Networks (PINN), we constructed a new, hybrid Deep Learning model that combines the strengths of GARCH with the flexibility of a Long Short-Term Memory (LSTM) Deep Neural Network (DNN), thus capturing and forecasting market volatility more accurately than either class of models are capable of on their own. We refer to this novel model as a GARCH-Informed Neural Network (GINN). When compared to other time series models, GINN showed superior out-of-sample prediction performance in terms of the Coefficient of Determination ($R^2$), Mean Squared Error (MSE), and Mean Absolute Error (MAE).

What problem does this paper attempt to address?

The problem this paper attempts to address is the prediction of volatility in financial markets. Specifically, the paper aims to develop a new hybrid deep learning model—GARCH-Informed Neural Network (GINN)—to improve the accuracy of market volatility predictions. Traditional GARCH models and their variants perform well in predicting stock volatility, but they underperform under certain market conditions, especially in capturing nonlinear market features. In recent years, deep learning models have shown high accuracy in time series prediction tasks but suffer from overfitting issues. Therefore, the paper constructs the GINN model by combining the advantages of the GARCH model with the flexibility of LSTM neural networks, aiming to overcome the limitations of single models and achieve more accurate market volatility predictions. The main contributions of the paper are: 1. **Model Innovation**: Proposes the GINN model, embedding the GARCH model as a regularization mechanism into the loss function to prevent overfitting, and utilizing the flexibility of LSTM neural networks to capture complex patterns of market volatility. 2. **Performance Evaluation**: Validates the predictive performance of the GINN model on multiple stock market indices by comparing it with GARCH, GJR-GARCH, TGARCH, and simple LSTM models. 3. **Parameter Optimization**: Determines the optimal weight values through parameter studies, ensuring the GINN model achieves optimal prediction accuracy. In summary, the goal of this paper is to develop a volatility prediction model that is both accurate and generalizable by combining classical statistical methods with modern machine learning techniques, thereby providing more valuable information for financial investment decisions.

GARCH-Informed Neural Networks for Volatility Prediction in Financial Markets

From GARCH to Neural Network for Volatility Forecast

A Hybrid GARCH and Deep Learning Method for Volatility Prediction

Volatility Prediction via Hybrid LSTM Models with GARCH Type Parameters

Combining Deep Learning and GARCH Models for Financial Volatility and Risk Forecasting

Comparing Deep Learning Models for the Task of Volatility Prediction Using Multivariate Data

Forecasting Daily Stock Volatility: A Comparison between GARCH and Recurrent Neuro-networks

A long short-term memory enhanced realized conditional heteroskedasticity model

Dynamic graph neural networks for enhanced volatility prediction in financial markets

Volatility forecasting using deep neural network with time-series feature embedding

Novel Volatility Forecasting Using Deep Learning–long Short Term Memory Recurrent Neural Networks

Sector Volatility Prediction Performance Using GARCH Models and Artificial Neural Networks

Support Vector Regression Based GARCH Model with Application to Forecasting Volatility of Financial Returns

Deep Learning Enhanced Realized GARCH

LSTM–GARCH Hybrid Model for the Prediction of Volatility in Cryptocurrency Portfolios

Gated deep neural networks for implied volatility surfaces

Risk Analysis of the Chinese Financial Market with the Application of a Novel Hybrid Volatility Prediction Model

Deep learning for volatility forecasting in asset management

A Hybrid Prediction Model Integrating GARCH Models With a Distribution Manipulation Strategy Based on LSTM Networks for Stock Market Volatility

Neural Generalised AutoRegressive Conditional Heteroskedasticity

Forecasting the Volatility of CSI 300 Index with a Hybrid Model of LSTM and Multiple GARCH Models