The Research of ARMA Model in CPI Time Series
The changing trend of CPI to a certain extent reflects the degree of inflation,which has a great significance on macro-control and research on national economic.ARMA model is one of the simple and practical models in financial time series analysis with relatively high forecast accuracy.The paper utilizing Eviews software,through the statistical analysis of CPI from the year of 1995 to 2008 monthly in China,through the ADF unit root test 1,by dint of the autocorrelation function ACF diagram2 and partial autocorrelation function PACF diagram3 to identify the model consequently establish the model,through the residual serial correlation test of the residuals of the model to select the correct model5.The predications of the model showed that the ARMA model is valid and forecast accuracy is relatively high
CPI time series ARMA ACF PACF
Fengwang ZHANG Wengang CHE Beibei XU Jingzhi XU
The School of Information Engineer and Automation Kunming University of Science and Technology Kunmi The School of Foreign Language and Literature Yunnan Normal University Kunming, China
国际会议
太原
英文
28-31
2013-04-06（万方平台首次上网日期，不代表论文的发表时间）