在2022年新型冠状病毒肺炎疫情在全国各地爆发，给人们的生活带来了许多困扰。为了分析疫情变化的规律，分析外部因素对疫情的影响，为以后的疫情提供预警、防控参考，本文使用2021年2月1日至2022年11月20日期间的每周新增新型冠状病毒肺炎感染病例数数据，对其建立干预模型，定量分析了干预变量如何对每周新增病例数变化趋势产生影响。

本文结合新闻和疫情数据特点，确立了四个干预变量：春运春节返工返校引起的大量人员流动、疫情管控措施、聚集性疫情事件和气温降低。因本文数据含有复杂的季节成分，文献常见的ARIMA模型拟合效果不佳，因此本文采用时间序列加法模型处理干预前的数据，本文所建最终模型的拟合效果达到98%。该研究结果对干预分析中的时间序列建模具有一定的参考意义。

In 2022, COVID-19 broke out all over the country, interrupting people's lives. In order to analyze the rule of epidemic change, analyze the impact of external factors, and provide early warning, prevention and control reference for future outbreaks, this paper uses the data of the number of new cases of COVID-19 infection per week from February 1, 2021 to November 20, 2022, to establish a intervention model, and quantitatively analyze how intervention variables affect the trend of the number of new cases per week.

This article establishes four intervention variables based on the characteristics of news and epidemic data: significant personnel turnover caused by returning to school during the Spring Festival, epidemic control measures, clustered epidemic events, and temperature reduction. Due to the complex seasonal components in the data in this article, the ARIMA model commonly used in literature has poor fitting performance. Therefore, this article uses a time series additive model to process the pre-intervention data, and the final model built in this article has a fitting effect of 98%. The research results have certain reference significance for time series modeling in intervention analysis.