副伤寒的求和自回归滑动平均模型预警应用实例

Application of the autoregressive integrated moving average model model for warning of paratyphoid

摘要: 目的　探讨时间序列求和自回归滑动平均模型（ARIMA）在副伤寒预警中应用的可行性。方法　利用SAS 9.0统计软件对《国家疾病报告管理信息系统》报告的杭州市副伤寒按周发病数进行ARIMA建模。结果　对副伤寒发病数序列建立三阶自回归模型AR（3），并绘制预警线图，对2007年7月发生的副伤寒暴发疫情进行了及时预警。结论　模型能够较好应用于副伤寒预警，为疫情防控提供了有力帮助。

Abstract: Objective　The study was conducted to evaluate the feasibility of the time series autoregressive integrated movingaverage model(ARIMA) model in early warning of paratyphoid. Methods　ARIMA modeling was established based on the weekly morbidity of paratyphoid in Hangzhou city reported by the National Disease Report Management System in SAS 9.0. Results　Athird order regression model AR (3) of the incidence sequence of paratyphoid was developed and the early warning curves plotted. As a result, an outbreak of paratyphoid in July 2007 was predicted in a timely manner. Conclusion　The model can be applied for early warning of paratyphoid, promoting the epidemic prevention and control.