Combination forecasting method and application based on the fractal dimension weight
Abstract
Since weights have an important impact on the prediction results in the combination forecasting, a combination forecasting method based on the fractal dimension weight was proposed to improve the predictive accuracy in this paper. According to the fitting error sequences of the single forecasting models, the reciprocal of box dimension normalization results were taken as the weights of single forecasting models. By using the weighted average method, the results of single forecasting models were combined. Through predicting the concentration of conventional pollutants DO, COD and NH3-N, it is shown that the prediction accuracy of the proposed method in this paper is higher than the other single forecasting models. It is verified that the proposed method in this paper is effective.
