Stock market forecasting model based on semi-parametric smoothing regression
Lingzhi Wang, Qin Fa-jin · Chinese Control Conference · 2012
In this paper, a novel Semi-parametric regression smoothing is presented for financial time series forecasting. Firstly, the Partial Least Square (PLS) technology is used to choosing and extracting the appropriate factors for these primary predictors from a number of economic variables. Secondly, the semi-parametric smooth regression with a penalized item is used to model for prediction, which GA is applied to search the optimal smoothing parameter in order to improve the smoothness of curve fitted. For testing purposes, this paper compare the new regression model's performance with some existing parametric regression model. Experimental results reveal that the predictions using the proposed approach are consistently better than those obtained using the other methods presented in this study in terms of the same measurements.