Prediction of water consumption in Beijing

Abstract

It applied the unbiased grey GM (1, 1) forecasting model and nonlinear prediction model to predict the water consumption of Beijing from the year 2001 to 2010. Because single prediction method lacks generalization, this paper combined unbiased grey GM (1, 1) predicting model with nonlinear prediction model called weighted combination model to predict the water consumption of Beijing. The results show that the error variance of the prediction result of the unbiased grey GM (1, 1) predicting model is small but the average absolute error of that is large, but the error variance of the result of nonlinear prediction model is large and the average absolute error of this is small. The weighted combination model can balance the above two kinds of prediction model, the mean absolute error and sum of squared errors are both between that of between two kinds of model. This weighted combination model makes the result more accurate and reliable and it can be used in the short term and the long term prediction of urban water consumption.