The Society of Convergence Knowledge Transactions ISSN:2287-8920(Print) 융복합지식학회논문지

The Consumer Price Index (CPI) is an important macroeconomic indicator used to measure changes in the prices of goods and services, and is utilized in economic trend analysis, as well as in the formulation and evaluation of government economic policies. In this study, we constructed and predicted an autoregressive error (ARE) time series model using data from January 2013 to December 2023 provided by the Bank of Korea Economic Statistics System. First, the autoregressive error time series model with a second-order trend was applied, and there were no problems with the significance of the regression coefficient and the explanatory power of the model. However, when conducting a residual analysis using the Portmanteau test to assess the model's fit, the results indicated the presence of autocorrelation. To improve the model, an autoregressive error time series model without a constant term including seasonality was applied to the second-order trend model. The estimation of regression coefficients using the backward elimination method showed significance at lags 1, 3, 4, and 9, and residual analysis indicated that the model was well-fitted, with no autocorrelation present at any lag. The consumer price index is predicted to rise from January to May 2024 and then fall from June.