A hybrid model based on support vector regression and modified harmony search algorithm in time series prediction

Abstract

Support vector regression (SVR) model has been widely applied to time series prediction. Due to the inherent non linearity and non-stationary characteristics of financial time series, conventional modeling techniques such as the Box-Jenkins autoregressive integrated moving average are not adequate for financial time series prediction. In this paper a hybrid model based on modified harmony search algorithm, and support vector regression (SVR) is proposed to predict financial time series. One of the problems in using support vector regression model is to determine the parameter values of SVR that in the proposed model, modified harmony search algorithm is used to optimize SVR parameters using search in the problem space finds the optimum values for each parameter. Then the optimized SVR is used to predict financial time series. The proposed method is tested on two sets of reliable financial datasets and experimental results on time series data show that the proposed model improved accuracy of prediction compared to other optimization methods.