Stock Market Risk Measurement Based on QGARCH and Machine Learning Algorithm

摘要

This study mainly uses the parametric and semi-parametric models to conduct financial risk measurement on four financial stock markets, including the Shanghai Composite Index, to help the regulatory authorities respond to financial market fluctuations in a timely manner. In this study, the traditional GARCH family model, the recursive quantile CAViaR model and the QGARCH model are applied to the stock market, and the machine learning algorithm is used to measure and analyze VaR. This paper finds that the QGARCH model has more advantages for financial market risk prediction, and it can reduce the fluctuation of the model and improve the stability of risk value by integrating with the machine learning model.