Education, Science, Technology, Innovation and Life
Open Access
Sign In

Research on the Trading Strategy Model Based on Dynamic Programming Algorithm

Download as PDF

DOI: 10.23977/csic2022.005

Author(s)

Haobo Sun, Xiaonan Dai, Kuotian Wei

Corresponding Author

Haobo Sun

ABSTRACT

This paper built three models to achieve daily purchase decisions and future property value estimates, the first is a dynamic planning purchase model, the second is the price grey prediction model, the third is the sale model. We take the forecast price of gold on the second day and the actual price on the first day as the starting point of analysis and use the ratio method to analyze and obtain the prediction coefficient. Then, the risk and yield in investing in gold and Bitcoin are initially taken as the starting point of analysis. It also adopts a mathematical method regarding the yield as a random variable and the risk as the coefficient representing the size of the yield fluctuation-variance. Weigh the return and risk of the portfolio by finding the Sharpe ratio is investigated. It is proven that our mathematical model provides the best strategy according to the Sharpe ratio calculations. Finally, we only change the value of trade cost in the case of controlling other condition variables unchanged. The control variable method is used to analyze the sensitivity of the strategy to trade costs. It is also continued to run our model when the trade costs increase or decrease by comparing the strategies given by the model under different trade costs under other conditions.

KEYWORDS

Grey prediction, Dynamic analysis, Sharpe ratio

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.