A Sense of Least Squares Support Vector Machine Chaotic Time Series Prediction Algorithm

Chaotic time series prediction is the chaotic time series prediction method of least square support vector for particle swarm optimization (PSO-LSSVM) to improve the precision of traditional time series prediction method. This method firstly uses the phase-space reconstruction method to reconstruct the time series samples and then use the particle swarm optimization to optimize the least square support vector machines, so as to obtain the optimal prediction model of chaotic time series. Finally, the classical chaotic time series Mackey-Glass is used to conduct simulation test on and analyze the optimal model. The simulation test results show that PSO-LSSVM accelerated the prediction speed, and improved the prediction precision. This indicated that PSO-LSSVM has application value in the chaotic time series prediction.