PARAMETER OPTIMIZATION METHOD FOR A NOVEL GREY RICCATI MODEL AND ITS APPLICATION

. The common grey prediction models include the GM(1,1) model, the GM(1,1) power model, the GM(1,N) model, the GM(N,1) model and so on, and they have different applicable scopes. The paper proposes a method to build a novel grey Riccati model and gives a general analytic expression of time response equation for simulation and predictions and the parameter optimization method of grey Riccati model. To improve the modeling precision of the grey Riccati model and the adaptability of data to the model, the paper improves the accumulated generating sequence of original data, i.e. making a new data transformation, and the parameters in its data transformation are simultaneously optimized with the generation coefficients of the 2 background values, which greatly improves the modeling accuracy. The paper builds the grey Riccati models for China's total electricity consumption and China's private car ownership. The results show that the grey Riccati models built with the method proposed have high simulation and prediction precision. The model proposed enriches the grey model system and has great significance for the in-depth research and generalized application of grey model.