The Prediction Model of Total Retail Sales of Social Consumer Goods of China

The autoregression,moving average,autoregression in moving average model were used to fit the total retail sales of social consumer goods from January 2002 to August 2008,and to establish the scattered coefficient model for establish forecasting the total retail sales of consumer goods.The model is used to predict the total retail sales of consumer goods from June 2008 to October 2008,and the total retail sales of consumer goods in 2009.The prediction errors of the model were estimated within 0.03,and the estimated values very closes to the real values,suggesting that the fitting effect can reach the expected purpose.