基于时间序列分析的广东省GDP的预测

The Forecast of GDP in Guangdong Province Based on Time Series Analysis

摘要: GDP是衡量一个国家或地区经济状况和发展水平的重要指标。因此，我们运用时间序列的方法对GDP变化作出预测是十分有意义的。在本文中，我们研究了ARIMA模型、指数平滑模型和方差倒数法组合预测模型三种模型对广东省2018~2021年GDP的预测效果。结果发现Holt两参数指数平滑模型的预测精度优于方差倒数法组合预测模型，组合预测模型优于ARIMA模型。

Abstract: GDP is an important indicator for measuring the economic status and development level of a country or region. Therefore, it is very meaningful for us to use the time series method to predict changes in GDP. In this article, we studied the predictive effects of three models, ARIMA model, exponential smoothing model, and reciprocal variance method combination prediction model, on the GDP of Guangdong Province from 2018 to 2021. The results showed that the prediction accuracy of the Holt two-parameter exponential smoothing model was superior to the reciprocal variance combined prediction model, and the combined prediction model was superior to the ARIMA model.