Forecast of Consumer Price Index-Take Beijing as an example

Abstract

: The consumer price index (CPI) reflects the relationship between the price changes of goods and services related to the life of residents and is an important indicator to evaluate the level of inflation. Because of the high randomness and volatility of CPI under the infectious diseases, it is very difficult to predict its trend accurately.In this paper, we combine the monthly CPI data of Beijing from January 2020 to July 2022, and use the ARIMA model, GM (1,1) and BP neural network model as the basis of the combined model to forecast the CPI of Beijing under the infectious diseases using two methods: ultra-short-term forecasting and conventional forecasting. It is obtained that the combined model has better forecasting effect than the single model, and the ultra-short-term forecasting effect is better than the conventional forecasting. Among them, the combination model using ARIMA-GM-BP for ultra-short-term forecasting is the best. Finally, the model and method were applied to forecast the CPI of Beijing in August as 102.079.