Research on Trading Strategy Selection Based on ARIMA and Linear Programming
Download as PDF
DOI: 10.23977/FEIM2022.030
Author(s)
Yuan Liu, Shiqin Zhou
Corresponding Author
Yuan Liu
ABSTRACT
The investment trading market is changing rapidly; how do the traders choose trading strategies in the highly uncertain trading market? Take gold and bitcoin as an example. Based on Bitcoin and Gold price data from September 11, 2016, to September 10, 2021, we used the ARIMA algorithm to predict the closing prices of bitcoin and gold in five years, R^2 is over 0.9 so that the predicted data are similar to the actual data. Then, we built a linear programming model with the maximization of return as the objective function and the variation of cash as the constraint while considering the risk factor. We proposed to split the feasible region to solve the linear programming model where the constraints contain absolute values. Moreover, finally, we obtained the strategy that could be provided for the trader, and the total value of the assets, in the end, is 42,051.42 USD.
KEYWORDS
ARIMA, Linear Programming, Trading Strategy, Risk Control