An improved differential evolutionary optimization algorithm for data-driven quantitative trading model

In this paper, we first develop a quantitative trading model based on data driven for simulating buy, sell and hold operations in stock trading. In order to optimize the investment returns of this model, this paper proposes an adaptive differential evolution algorithm incorporating the Q-learning algorithm, called the DE algorithm. The method dynamically adjusts the key parameters F and CR of the algorithm through the interactive learning of the Q-learning algorithm, so as to improve the adaptability and performance of the algorithm in different optimization stages. The experimental results show that the improved algorithm can effectively solve the optimization problem based on quantitative trading models and improve the investment returns in real stock trading, which verifies the potential of the method to be applied in complex financial environments.