基于ARIMA模型的传染病疫情计量研究

On the Measurement of Infectious Diseases based on ARIMA Model

通过自回归移动平均模型(ARIMA)曲线拟合乙型肝炎疫情的发病数,以此为依据对传染病行业的需求进行分析和预测。选择2009~2018年乙型肝炎发病数据,通过时间序列分析法进行处理,采用序列平稳化、模型估计、模型检验拟定ARIMA季节性模型。并使用2013年1月-2016年12月我国乙型肝炎发病数据来进行模型验证,再预测未来一年的乙型肝炎月发病数,可得知2009-2018年乙型肝炎的发病数据可以被ARIMA(3,1,2)(1,1,1)12模型很好地拟合。根据模型预测得到,2019年我国乙型肝炎的发病数为1224591例,总体呈上升趋势,在3、7、8月份逐月增加,随后又有所减少。

This paper analyzes and predicts the demand of infectious disease through the incidence of hepatitis b epidemic fitted by the curve of the Autoregressive Integrated Moving Average Model(ARIMA). The incidence data of hepatitis b from 2009 to 2018 were selected and processed by time series analysis. The ARIMA seasonal model was developed by using sequence stabilization, model evaluation and model test. The data of hepatitis b incidence in China from January 2013 to December 2016 were used for model validation to predict the monthly incidence of hepatitis b in the next year. It is known that incidence data of hepatitis b in 2009-2018 can be well fitted by ARIMA model(3,1,2)(1,1,1). According to the prediction of the model, the incidence of hepatitis b in China in 2019 is 1224,591 cases, with an overall upward trend, increased in March, July and August, then decreased.