Welcome to Francis Academic Press

Academic Journal of Business & Management, 2022, 4(4); doi: 10.25236/AJBM.2022.040418.

Optimization of Portfolio Investment Strategy Based on Markov Decision Process Model

Author(s)

Xiangzhao Hao, Runxing Jia

Corresponding Author:
Xiangzhao Hao
Affiliation(s)

Tianjin university, Tianjin, 300192, China

Abstract

With the continuous development of the world financial market, investors are often faced with a wide variety of assets but do not know how to choose. In order to obtain the maximum benefits, how to choose the optimal investment strategy has become a top priority. In this paper, based on Markov decision process model, we study the optimal strategy for investment with two important assets in the market, gold and Bitcoin. we solve the problem with a Markov decision process model and measure the risk of the trade by financial indicators such as deviation rates and bullish indicators. Then a difference-in-difference formula for the asset appreciation rate is introduced to optimize the model by incorporating the long-term impact of trading on assets. Also based on the iterative algorithm of the sample path, an iterative algorithm of the strategy that can be applied online is designed, and it was finally calculated that after five years of investment using this strategy under the initial condition of $1000, $348287.46 could be obtained.

Keywords

Markov decision process model; Portfolio investment strategy; Proportional transaction costs

Cite This Paper

Xiangzhao Hao, Runxing Jia. Optimization of Portfolio Investment Strategy Based on Markov Decision Process Model. Academic Journal of Business & Management (2022) Vol. 4, Issue 4: 89-95. https://doi.org/10.25236/AJBM.2022.040418.

References

[1] Taksar M, Klass M J, Assaf D. A diffusion model for optimal portfolio selection in the presence of brokerage fees [J]. Mathematics of Operations Research, 1988, 13(2): 277-294. 

[2] Huang Yonghao, Chen Xi. Investment Portfolio Optimization with Proportional Fees Based on Sensitivity Analysis [J]. Control and Decision, 2014, 29(07): 11811186.

[3] Meng Xianghuan. Research on the application of risk portfolio investment model based on multiobjective planning [D]. Wuhan University of Science and Technology, 2013.

[4] Cao X R. Stochastic learning and optimization: A sensitivity-based approach [M]. New York: Springer, 2007: 341-380.

[5] PaperWeekly, When Deep Learning Meets Quantitative Trading - Models, https://blog.csdn.net/ c9Yv2cf9I06K2A9E/article/details/106561139, [2021-2-19]

[6] Meng Xianghuan. Research on the application of risk portfolio investment model based on multi-objective planning [D]. Wuhan University of Science and Technology, 2013.

[7] Huang Wenlan. Research on Portfolio Optimization Based on Multi-Objective Evolutionary Algorithm [D]. Harbin University of Commerce, 2020. DOI: 10.27787/d.cnki.ghrbs.2020.000514.