A tunable epsilon-tube in support vector regression for refining parameters of GM(1,1 | τ) prediction model - SVRGM(1,1 | τ) approach

This paper introduces a novel SVRGM(1,1 | τ) prediction model for forecasting economic indexes like stock price indexes or future trading indexes. SVRGM(1,1 | τ) model employ the support vector regression (SVR) learning algorithm to improve the control and environment parameters in grey model GM(1,1| τ) model, that is, enhancing generalization capability in the non-periodic short-term prediction. Therefore, this proposed method could smooth the overshooting problem, that often occurred in GM(1,1| τ) model or autoregressive moving-average (ARMA) method, so as to achieve better the prediction accuracy.