Welcome to Francis Academic Press

International Journal of Frontiers in Sociology, 2025, 7(6); doi: 10.25236/IJFS.2025.070603.

The Dynamic Prediction of Market Risk Based on the GARCH Model for the CSI 500 Index

Author(s)

Zeyu Fan, Aihua Wang

Corresponding Author:
Aihua Wang
Affiliation(s)

College of Business and Economics, Shanghai Business School, Shanghai, China

Abstract

This study aims to dynamically predict the market risk of the CSI 500 Index using the GARCH model. After analyzing the data, it is found that the daily return series of the CSI 500 Index does not follow a normal distribution, prompting the selection of a GARCH model with a student’s t-distribution. The GARCH (1,1)-t model is identified as the optimal model through fitting and diagnostic testing. The model is then used for dynamic market risk forecasting, with Value at Risk (VaR) applied for risk assessment from March 2024 to March 2025. The model's reliability is verified through Kupiec’s likelihood ratio backtesting, evaluating VaR exception frequencies at multiple confidence levels. Based on the findings, the study offers recommendations: investors should strengthen risk awareness, use derivatives for hedging, and adjust investment strategies based on the GARCH model's volatility predictions; financial institutions should develop intelligent risk management tools and optimize asset allocation; and regulators should enhance dynamic risk monitoring and promote the digital transformation of financial institutions.

Keywords

Financing GARCH Model, CSI 500 Index, Market Risk, Dynamic Forecasting

Cite This Paper

Zeyu Fan, Aihua Wang. The Dynamic Prediction of Market Risk Based on the GARCH Model for the CSI 500 Index. International Journal of Frontiers in Sociology (2025), Vol. 7, Issue 6: 13-20. https://doi.org/10.25236/IJFS.2025.070603.

References

[1] Jiang Min, Zhang Chuyi, Sun Deshan. Research on Stock Price Prediction Based on the PCA-GARCH-LSTM Model [J]. Software Guide,2025.24(01).

[2] Chen Yanshan, Zhang Xingfa, Tian Yue, et al. A Portmanteau Q Test for Maximum Likelihood Estimation of the GARCH Model Based on High-Frequency Data [J]. Journal of Guangzhou University (Natural Science Edition), 2024, 23(05).

[3] Chen Zhuming, Zhang Xiangyu. Research on the Hedging Role of Stablecoins in Cryptocurrency Market Volatility—Based on the DCC-GARCH Dynamic Model [J]. Modern Financial Research, 2024, 29(10).

[4] Wang Yifan, Wang Shuting, Wang Jia. Research on the Cross-Market Risk Linkage and Spillover Effects of Green Bonds under the "Dual Carbon" Background [J]. Business Exhibition Economics, 2024, (19).