随机时间序列分析法在传染病预测中的应用

Application in infectious disease forecasting by ARIMA model

目的阐述ARIMA模型拟合时间序列的方法和步骤,并将其应用于乙型肝炎的预测,为传染病预警系统提供决策依据。方法利用SPSS统计软件求解适宜的ARIMA模型,据所得误差评价预测效果。结果通过对乙型肝炎发病率的预测,相对误差在15%左右,预测效果较为可靠。结论在乙型肝炎的近期预测中引入时间序列的ARIMA模型分析方法,能够对传染病的预测产生积极的指导意义。

Objective The approach and procedure to fit time series with ARIMA models are discussed briefly. The application to forecast hepatitis B is given to help infectious diseases forecasting system. Methods Proper ARIMA model is obtained with SPSS system and the effectiveness is evaluated. Results The error of prediction to hepatitis B is around 15%, which show a satisfactory effectiveness. Conclusion It is both necessary and practical to apply the approach of ARIMA model in fitting time series to predict hepatitis B with a short lead time.