Application of support vector regression trained by particle swarm optimization in warrant price prediction

Warrant price prediction is very important to investment. Support vector regression technique is a learning procedure based on statistical learning theory, which employs the training data to build an excellent forecasting model in the situations of small sample. The prediction ability of support vector regression is influenced by its training parameters. Particle swarm optimization is applied to choose the parameters of support vector regression. Then, support vector machine trained by particle swarm optimization is presented to predict warrant price. The prediction ability of warrant price of the method is studied by the historical warrant price data including seven data points of a certain warrant. It can be seen that the warrant price prediction performance of PSO-SVR is better than that of BPNN by the experimental results.