Multivariate Nonlinear Prediction of Shenzhen Stock Price

Abstract

In this paper, an attempt is made to predict stock price movement on Shenzhen stock market of China with nonlinear dynamical theory. Multivariate nonlinear prediction method based on multidimensional phase space reconstruction is considered. We propose a multivariate nonlinear model in forecasting stock price, and compare the prediction accuracy of our model with univariate nonlinear prediction model. The results show that multivariate nonlinear prediction model outperforms univariate nonlinear prediction model. Multivariate nonlinear prediction model is a useful tool for stock price prediction in emerging markets.