Stock Price forecasting based on Quantum Particle swarm optimization
The stock market is very volatile, so the change of the stock price is also widely concerned by investors. In this paper, a new stock price forecasting model based on Quantum Particle Swarm Optimization(QPSO) , Quantum Bee Colony Optimization Algorithm(QABC) and Quantum Fruit Fly Optimization Algorithm (QFOA) is proposed. The three methods all use BP neural network to adjust the parameters of particle swarm, bee colony and Drosophila to reach the optimal parameters. Taking the daily closing price of CITIC Securities and Tianfeng Securities, a large-scale and a small-scale securities company, as the object of empirical analysis, comparing the accuracy of the three methods in predicting stocks, it also analyzes whether the size of the company has an effect on the accuracy of the model. The results show that the prediction effect of qpso is the best, and the size of the company has some influence on the prediction effect.