Research on Gold Bitcoin Synthesis Algorithm Based on Particle Swarm Optimization and Dynamic Programming

In this paper, we hope to study the transaction process of bitcoin and gold, two unstable assets, to find a trading model that can predict as accurately as possible and put forward a good algorithm. In the beginning, this paper predicted the future price movements of gold and bitcoin based on the long short-term memory neural network model. Then, based on the price of the day, this paper established a strategy model based on the Sharpe ratio and particle swarm algorithm. In order to better quantify the relationship between daily investment risks and returns, we introduced the Sharpe ratio and set the objective function to adjust the position and the value of the Sharpe ratio on the third day when the next three days are unchanged when the objective function is at the maximum value is the situation with the maximum risk and benefit. Finally, this paper established a comprehensive planning model for cash, gold, and Bitcoin through the dynamic programming process. A particle swarm algorithm simulated the optimization process, and the optimal solution was found. We obtained the Sharpe rate on the third day as the objective function.