| The Consumer Price Index(CPI) is a widely used measurement of cost of living. It not only affects the government monetary, fiscal, consumption, prices, wages, social security, but also closely relates to the residents’ daily life. As an indicator of inflation in China economy, the change of CPI undergoes intense scrutiny. Therefore, precisely forecasting the change of CPI is significant to many aspects of economics, some examples include fiscal policy, financial markets and productivity. Also, building a stable and accurate model to forecast the CPI will have great significance for the public, policymakers and research scholars. The time series model is widely used in study the CPI in the early time. However, the forecasting results are not satisfactory enough since the financial time series are usually difficult to satisfy the stability assumptions. With the widely application of the artificial intelligence model, scholars introduce it to the CPI prediction research later. In the modeling process of CPI forecasting, both the noise and the selection of model parameters will influence the accuracy of prediction. In order to solve these two problems, this article comes up with three kinds of models including EEMD-PSO-BP, EEMD-PSO-WNN and EEMD-PSO-SVR which are based on the EEMD de-noising algorithm and PSO optimization algorithm. Using three kinds of new models, respectively, to conduct multi-step prediction research on CPI data from January 1995 to August 2015, the experimental results show that EEMD algorithm and PSO algorithm can improve the performance of the original model in different degree. According to the contrast of the prediction result of three kinds of models, it is concluded that performance of the EEMD-PSO-SVR is best.This model can be applied to the prediction research of the consumer price index. |
