Year: 2025
Type: article
Abstract: With the rapid development of financial market in China, systematic risk control has become an important research topic in securities investment. In recent years, financial markets have experienced frequent and significant fluctuations around the world, bringing unprecedented challenges to investors, especially in the prevention and control of systemic risks. Systemic risk, i.e. market risk, is the risk that affects all assets and cannot be completely eliminated by diversifying the investment. In order to deal with this kind of risk, based on the financial mathematical modeling method, this paper constructs a comprehensive risk control model for securities market investment, in order to help investors and regulators to manage risks more scientifically and effectively, and to improve the stability of investment and the rationality of return. In the aspect of systematic risk prediction, this paper introduces machine learning technology to construct a systematic risk model with stronger prediction ability. By selecting multiple risk factors such as market volatility (MVI), liquidity (LSI), linkage (MCI) and volume pressure (VPI), a multi-factor prediction model is established, which uses machine learning algorithms such as XGBoost and stochastic forest to predict the change trend of future systematic risk. During the training process, the model improves the ability to capture market risks through feature engineering and parameter optimization, so that it can more effectively predict the possible market fluctuations in the future. At the same time, this paper makes a feature importance analysis on the model, which helps to identify the factors that have a key impact on the systematic risk and provides further basis for the application and improvement of the model. On the whole, the model proposed in this paper has high practicability and algorithm efficiency while effectively controlling the risks, which provides a scientific basis for the risk control in the securities market.
DOI: 10.54097/yresac18
Authors: Chang Liu, Rongjie Cai, Yuting Deng, Guanliang Chen, Yuqi Shan, Min Luo, Xin Guo, Chuxiao Ye
Institution: Guangdong Ocean University
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4409787782