Optimized GM(1,1)model and application based on least square method
In order to more accurately capture and predict economic trends,especially in the field of urban transportation,this paper is based on the classic GM(1,1)model,uses the least squares method to reconstruct sample data and background value weights,optimizes and innovates the GM(1,1)model,and proposes an optimized GM(1,1)model based on the least squares method,and applies the model to the data fitting and prediction of the urban transportation indicator"private car ownership".It can be seen from the specific example that the average relative error of the fitted value is reduced from 1.73％to 0.4％,and the average relative error of the predicted value is reduced from 9.94％to 3.85％.The results show that,under the premise of meeting the quasi exponential conditions of grey modeling,the new model algorithm can achieve the optimal state of the fitting sequence in the least squares sense,which can provide certain technical support for research and application in related fields.
GM(1,1)modelleast square methodprivate car ownership
万方数据
维普
