Stock Return Forecast with LS-SVM and Particle Swarm Optimization

Stock return forecast has been an important issue and difficult task for both shareholders and financial professionals. To tackle this problem, we introduce least square support vector machine (LS-SVM), an improved algorithm that regresses faster than standard SVM, and dynamic inertia weight particle swarm optimization (W-PSO), that outperform standard PSO in parameter selection. The work of this paper is as following: First, forecast daily stock Return of Shanghai Security Exchanges of China using Back Propagation Neural Network (BPNN) and LS-SVM. Secondly, forecast the stock return using LS-SVM optimized by W- PSO. Finally, make a comparative analysis of the three algorithms. We reached conclusion that, in terms of forecast accuracy, LS-SVM outperforms BPNN, and when LS-SVM is optimized by W-PSO, the best result is achieved.