Chines Journal of Vector Biology and Control ›› 2014, Vol. 25 ›› Issue (3): 231-234.DOI: 10.11853/j.issn.1003.4692.2014.03.010

Previous Articles     Next Articles

Application of autoregressive integrated moving average model in prediction of hemorrhagic fever with renal syndrome

WEI Ya-mei1, GUO Na-na2, HAN Xu1, HAN Zhan-ying1, ZHANG Yan-bo1, QI Shun-xiang1, LI Qi1   

  1. 1 Hebei Center for Disease Control and Prevention, Shijiazhuang 050021, Hebei Province, China;
    2 Handan Center for Disease Control and Prevention
  • Received:2014-01-10 Online:2014-06-20 Published:2014-06-20
  • Contact: LI Qi, Email: liqinew@126.com

差分自回归移动平均模型在肾综合征出血热发病预测中的应用研究

魏亚梅1, 郭娜娜2, 韩旭1, 韩占英1, 张艳波1, 齐顺祥1, 李琦1   

  1. 1 河北省疾病预防控制中心病毒病防治所, 石家庄 050021;
    2 邯郸市疾病预防控制中心
  • 通讯作者: 李琦, Email: liqinew@126.com
  • 作者简介:魏亚梅, 女, 主管医师, 主要从事自然疫源性疾病防制工作。
  • 基金资助:

    河北省自然科学基金(C2007000944)

Abstract:

Objective To evaluate the application of autoregressive integrated moving average (ARIMA) model in the prediction of monthly incidence of hemorrhagic fever with renal syndrome (HFRS). Methods The database of monthly epidemic situation of HFRS in Hebei province, China from 1986 to 2011 was constructed with SPSS 16.0. A mathematic model was constructed using ARIMA of SPSS 16.0 and used to predict the epidemic situation in 2012. Results The HFRS incidence presented obvious seasonal periodicity during 1986 to 2011 in Hebei province. ARIMA (0, 1, 1)×(0, 1, 2)12 model best fitted the incidence of HFRS from January 1986 to December 2011. The actual average incidence of HFRS in 2012 fell within the 95% confidence interval of prediction. Conclusion ARIMA model fits well in the prediction of HFRS incidence, and is suitable for use in epidemiological prediction to provide a basis for the prevention and control of HFRS.

Key words: Hemorrhagic fever with renal syndrome, Autoregressive integrated moving average model, Prediction

摘要:

目的 探讨差分自回归移动平均模型(ARIMA)在肾综合征出血热(HFRS)预测分析中的应用。方法 用SPSS 16.0软件建立1986-2011年河北省HFRS逐月疫情资料数据库, 用ARIMA相关模块进行建模拟合并进行预测分析。结果 河北省1986-2011年HFRS发病数呈现明显的季节周期性。筛选ARIMA(0, 1, 1)×(0, 1, 2)12模型为最优模型, 对河北省2012年各月发病数进行预测, 2012年1-12月实际值均落入了预测值的95%可信区间内。结论 ARIMA模型可以很好地拟合HFRS发病数的变动趋势, 并可用于预测未来疫情, 为HFRS防控工作提供依据。

关键词: 肾综合征出血热, 差分自回归移动平均模型, 预测

CLC Number: