The Consumer Price Index Forecast Based on ARIMA Model

Abstract

Inflation forecasts becomes a key input of monetary policy decision. CPI is a measure of inflation, however, an important economic indicator. Based on the monthly CPI data from January 2000 to December 2009, the thesis firstly statistically indentifies the correlation function and the partial correlation function of consumer price index, tests the stationarity of ADF, then uses ARIMA model to test residual serial autocorrelation, lastly makes a short-term estimation on monthly CPI of our country in 2010. Empirical results show that ARIMA (12,1,12) model provides a better prediction for the monthly consumer price index (CPI) of our country in 2010. CPI forecast based on the results of the Government formulating appropriate monetary policy