Combination Forecasting Model Using Grey Verhulst Models Coupling to Regression Analysis

Firstly, grey verhulst model based on fractional order accumulate is deduced. Then, take advantage of differential and difference to find the corresponding relationship between the grey verhulst model of each order and the regression equation. Giving the level of significance a, carry out the significance test for the regression equations. Picking up the orders of grey verhulst models corresponding to the first two regression equations which are best-fit for the original time series, then establish the grey verhulst models which have the orders picked up before. Lastly, using these two grey verhulst models to establish the weight arithmetic average combination forecasting model based on grey associate analysis. The practical example indicates that the fitting precision of our model is improved significantly comparing with single forecasting models, logistic model and fractional order accumulate grey verhulst model. It is proved to be an available and practical method for forecasting.