Stock market forecast based on wavelet neural network optimized by Cuckoo search

摘要

As a typical nonlinear deterministic dynamical system, stock market can be predicted by Wavelet Neural Network. Since Cuckoo Search is a new heuristic bionic group intelligent optimization algorithm, it can be widely used in various optimization problems. In this paper, we use Cuckoo Search (CS) to optimize the initial parameters of wavelet neural network. Results of the experiment show that the optimized CS-WNN has higher prediction accuracy than the traditional WNN in stock market forecast.