Abstract:

Forecasting longterm intensity of carbon emissions in China has important significance for policy makers to macroeconomy management and EnergySaving Emission Reduction Efficiency.Based on analysis of the autoregressive integrated moving average (ARIMA) and neural networks (NN) models,this paper presented an ensemble approach to the intensity of carbon emission time series forecasting which integrated ARIMA with NN.The time series was considered of a linear autocorrelation structure and nonlinear structure,and then the change trend of intensity of carbon emissions was analyzed and predicted.The forecast results indicate that the carbon intensity is gradually declined in the next ten years，but the intensity of  carbon emissions  in 2020 decreases only by 34% based on the 2005 level.The Chinese government proclaimed a mitigation target which proposed that intensity of  carbon emissions in 2020 would be reduced by 40%-45% based on the 2005 level.Therefore,we have to adjust macroeconomic policy and take all kinds of policy measures to achieve the goal.