Complex System Analysis on Voter Stochastic System and Jump Time Effective Neural Network of Stock Market

Abstract

The finite-range voter system, one of stochastic particle systems, is applied to model a financial price process for further description and investigation of fluctuations of Shanghai Composite Index. For different parameter values of the intensity λ and the range R, we investigate the statistical behaviors of the simulation data for this financial model. Then we develop the random jump time effective neural network model to forecast the fluctuations of Shanghai Composite Index. Moreover, we compare the two models by analyzing the returns and the absolute returns of Shanghai Composite Index, the simulation data and the predictive data through Detrended Fluctuation Analysis and classic Rescaled Range Analysis.