Proceedings of the 3rd International Conference on Big Data Economy and Digital Management, BDEDM 2024, January 12–14, 2024, Ningbo, China

Research Article

Optimization Analysis of Portfolio Method Based on High Order Moment

Download40 downloads
  • @INPROCEEDINGS{10.4108/eai.12-1-2024.2347289,
        author={Kai  Shi},
        title={Optimization Analysis of Portfolio Method Based on High Order Moment},
        proceedings={Proceedings of the 3rd International Conference on Big Data Economy and Digital Management, BDEDM 2024, January 12--14, 2024, Ningbo, China},
        publisher={EAI},
        proceedings_a={BDEDM},
        year={2024},
        month={6},
        keywords={high moment portfolio investment parametric strategy b-s-k modle bayesian optimization},
        doi={10.4108/eai.12-1-2024.2347289}
    }
    
  • Kai Shi
    Year: 2024
    Optimization Analysis of Portfolio Method Based on High Order Moment
    BDEDM
    EAI
    DOI: 10.4108/eai.12-1-2024.2347289
Kai Shi1,*
  • 1: Shandong Technology and Buisiness University
*Contact email: 3538598297@qq.com

Abstract

In this paper, a dynamic high order moment parametric portfolio investment decision model (B-S-K) is proposed to solve the deficiency of risk measurement model in existing high order moment portfolio investment models. This model uses MIDAS-QR model and parametric portfolio investment strategy to improve the timeliness, accuracy and robustness of risk measurement, reduce the number of parameters to be estimated and improve the solving efficiency of the model. In the empirical study, the model is applied to the individual stock and industry index of the Chinese stock market, and the results show that the model has advantages in risk measurement and portfolio investment decision. Specifically, the dynamic high-order moment risk measure based on MIDAS-QR model takes into account the time-varying characteristics of financial risk and has little influence on outliers. In addition, P/E ratio, book value ratio and dynamic skewness risk are positively correlated with portfolio weight, while conditional volatility and dynamic kurtosis risk are negatively correlated with portfolio weight, which provides explanations for portfolio investment decisions. Compared with other models, B-S-K model shows advantages in terms of return, risk and risk-adjusted return. In short, B-S-K model improves the risk measurement and portfolio investment decision through MIDAS-QR model and parametric portfolio investment strategy, and has better performance and effect.