Research on China’s GDP Forecast - Taking the ARIMA Model as an Example

Zhejiang University of Science and Technology, 310023 Hangzhou, Zhejiang Province, China

^* Corresponding author: 1220662017@zust.edu.cn

Abstract

Gross Domestic Product (GDP) serves as an important indicator for measuring a country’s economic situation. By predicting and analyzing GDP, one can understand the economic performance of the region in the coming period, especially in the context of economic transformation and frequent external shocks. Accurate GDP predictions are of great significance for preventing economic risks, optimizing resource allocation, and formulating other policies. Most existing studies on ARIMA models focus on low-frequency data from developed countries, while there is a lack of research on developing countries. This study is based on the ARIMA (autoregressive integrated moving average) model and explores its applicability in China’s GDP forecasting. Using annual data from 1960 to 2024 in China, the ARIMA model is constructed using the Box-Jenkins method, parameters are optimized through ADF test, AIC criterion, and out-of-sample test, and prediction accuracy is evaluated using RMSE and MAE. The results show that the ARIMA (3,2,0) model has a good predictive effect. The predicted growth rate of GDP in the next five years is 4.5%. Based on the research results, it can provide high-precision short-term prediction tools for government departments and expand the application scope of time series models in emerging economies.