A Fusion of Statistical and Machine Learning Methods: GARCH-XGBoost for Improved Volatility Modelling of the JSE Top40 Index
Preprint
- 4 August 2025
- preprint
- Published by MDPI AG in Preprints.org
Abstract
Volatility modelling is a key feature of financial risk management, portfolio optimisation, and forecasting, particularly for market indices such as the JSE Top40 Index, which serves as a benchmark for the South African stock market. This study investigates volatility modelling of the JSE Top40 Index log-returns from 2011 to 2025 using a hybrid approach that integrates statistical and machine learning techniques through a two-step approach. The ARMA(3,2) model was chosen as the optimal mean model, using the \texttt{auto.arima()} function from the \texttt{forecast} \texttt{package} in \textts{R} (version 4.4.0). Several alternative variants of GARCH models, including sGARCH(1,1), GJR-GARCH(1,1), and EGARCH(1,1), were fitted under various conditional error distributions (i.e., STD, SSTD, GED, SGED, and GHD). The choice of the model was based on AIC, BIC, HQIC, and LL evaluation criteria, and ARMA(3,2)-EGARCH(1,1) was the best model according to the lowest evaluation criteria. Residual diagnostic results indicated that the model adequately captured autocorrelation, conditional heteroskedasticity, and asymmetry in JSE Top40 log-returns. Volatility persistence was also detected, confirming the persistence attributes of financial volatility. Thereafter, the ARMA(3,2)-EGARCH(1,1) model was coupled with XGBoost using standardised residuals extracted from ARMA(3,2)-EGARCH(1,1) as lagged features. The data was split into training (60ARMA(3,2), EGARCH(1,1), Forecasting, Hybrid model, JSE Top40 Index, Machine Learning, Risk Management, Time Series, Volatility Modelling, XGBoost.%), testing (20%), and calibration (20%) sets. Based on the lowest values of forecast accuracy measures (i.e., MASE, RMSE, MAE, MAPE, and sMAPE), along with prediction intervals and their evaluation metrics (i.e., PICP, PINAW, PIWA, and PINAD), the hybrid model captured residual nonlinearities left by the standalone ARMA(3,2)-EGARCH(1,1) and demonstrated improved forecasting accuracy. This highlights the robustness and suitability of the hybrid ARMA(3,2)-EGARCH(1,1)-XGBoost model for financial risk management in emerging markets and signifies the strengths of integrating statistical and machine learning methods in financial time series modelling.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Futures volatility forecasting based on big data analytics with incorporating an order imbalance effectInternational Review of Financial Analysis, 2022
- Modelling stock returns volatility with dynamic conditional score models and random shiftsFinance Research Letters, 2022
- Risks in emerging markets equities: Time-varying versus spatial risk analysisPhysica A: Statistical Mechanics and its Applications, 2020
- New HEAVY Models for Fat-Tailed Realized Covariances and ReturnsJournal of Business & Economic Statistics, 2017
- Subjectively Interesting Component Analysis2016
- Volatility forecast using hybrid Neural Network modelsExpert Systems with Applications, 2014
- Dynamic Models for Volatility and Heavy Tails2013
- Analysis of Financial Time Series2010
- On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on StocksThe Journal of Finance, 1993
- Generalized autoregressive conditional heteroskedasticityJournal of Econometrics, 1986