A Hybrid Model Combined Grey Prediction and Autoregressive Integrated Moving Average Model for Talent Prediction

Chuen-Jiuan Jane

《Journal of Grey System》 21卷2期 (2018/06) Pp. 91-102

摘要

Talent demand forecast is the basis for economic growth which is also a prerequisite for the discipline optimization of higher education. In order to obtain the total demand forecast disciplinary talents, this paper proposes an optimal model which establishes a adjustment mechanism from the regulative view of the economic and social needs. First, a corresponding prediction index system is obtained based on the scale factors for extraction of talent demand Prediction. Then, a linear and nonlinear hybrid model with predictive ability is established for the complexity of the factors that influence the demand for talent BP neural network simulation curve, extract the nonlinear factors, the use of GM-ARIMA model residuals generated by nonlinear model fitting realize the benefits of a variety of complementary model to improve the talent scale of demand Prediction accuracy. Finally, according to the feasibility of the model scale discipline structure Talent Demand in China is predicted to prove the prediction process to predict the results of effectiveness.

關鍵字

BPNN ； GM-ARIMA ； Residual Compensation ； Talent Prediction