ARIMA乘积季节模型在我国甲肝发病预测中应用.docVIP

ARIMA乘积季节模型在我国甲肝发病预测中的应用 王 超1，丁 勇2，陆 群3，吴 静2* （1南京医科大学生物医学工程系，2数学与计算机教研室，3第一临床医学院，江苏 南京 210029） [摘 要] 目的：应用ARIMA乘积季节模型对我国病毒性甲型肝炎进行预测分析，为甲型肝炎的防治提供决策依据。方法：对1994～2012年我国甲型肝炎月发病数的历史疫情数据建立ARIMA乘积季节模型，应用Eviews 6.0软件进行模型拟合，对2013年上半年甲型肝炎的月发病数进行预测，并用实际数据评估模型预测效果。结果： ARIMA(1,1,0)(2,l,2)12模型较好地拟合了既往甲肝的实际发病序列，也获得了较好的预测效果。结论：ARlMA模型能够较好地模拟我国甲型肝炎的发病趋势，预测效果良好，可为甲肝疫情的防控提供一定的科学数据。 [关键词] ARIMA乘积季节模型；时间序列；甲肝；预测 [中图分类号] R512.6 Application of multiple seasonal ARIMA model in forecasting the incidence of hepatitis A in China Wang Chao1,Ding Yong2,Lu Qun3,Wu Jing2* (1Department of Biomedical Engineering, 2Department of Mathematics and Computer, 3The First Clinical College,NJMU,Nanjing 210029,China ) [Abstract] Objective:To forecast the incidence of hepatitis A in China by multiple seasonal ARIMA model, and to provide decision basis for prevention and control of hepatitis A.Methods: ARIMA model was fitted with data of monthly reported cases in China from Jan.1994 to Dec.2012 for hepatitis A. Eviews 6.0 software was performed to construct the ARIMA model, and the constructed model was applied to predict the monthly incidence in the first part of 2013．The model was evaluated by actual data.Results: ARIMA(1,1,0)(2,l,2)12 exactly fitted the incidence of the previous time series，and got a good result on the predictive incidence in 2013．Conclusion: The multiple seasonal ARIMA model can be used to make a short-term prediction and a dynamic analysis on the incidence of hepatitis A in China,and can provide a scientific basis for the prevention and control of the epidemic． [Key words] multiple seasonal ARIMA model;Time series; Hepatitis A; Prediction 甲型病毒性肝炎，简称甲型肝炎、甲肝，是由甲型肝炎病毒(HAV)引起的、以肝脏炎症病变为主的传染病，主要通过粪-口途径传播。粪口传播的方式是多样的，一般情况下，日常生活接触传播是散发性发病的主要传播方式，因此在集体单位如托幼机构、学校和部队中甲型肝炎发病率较高。水和食物的传播，特别是水生贝类如毛蚶等是甲型肝炎爆发流行的主要传播方式。例如1988年春季的上海甲肝大爆发，主要就是由市民食用受到甲肝病毒污染的毛蚶引起的，此次疫情波及面极广，短短4个月内罹患人数达到31万之多，造成了极大的社会恐慌。甲肝患病人群遍及各年龄段，主要为儿童和青少年。成人甲肝的临床症状一般较儿童重。冬春季常是甲肝发病的高峰期。自上海甲肝大爆发后，政府和卫生部门对甲肝的防控形势十