Quantitative Trading Method based on Neural Network Machine Learning

Abstract

Quantitative trading plays an essential role in the investment field with its advanced mathematical models for computer-aided trading of investment strategies. The artificial neural network algorithm is the trading algorithm with the largest amount of funds managed in the world. Due to the short history of quantitative trading research in China, large-scale funds have not been reported to be managed by the neural network algorithm. The results of tests on financial derivatives using neural networks with different structures demonstrate that the neural network strategies all have positive expected return. Within a considerable range of changes in structure. In this paper, the python language is majorly used to design a model implementation plan for a quantitative trading system reading currently widely recognized stock technical indicators, such as MA, MACD, KDJ, and BOLL. Additionally, position management strategies are optimized. Furthermore, a quantitative trading method based on neural network machine learning is constructed and verified with examples.