SARIMA模型在戊型肝炎发病趋势预警预测中的应用

发布时间：2018-06-17 01:35

本文选题：季节时间序列模型 + 戊肝　；参考：《中国人兽共患病学报》2015年02期

【摘要】：目的采用SARIMA模型对福建省戊肝发病趋势进行预测,为预警和风险评估提供定量数据。方法以Eviews 5.0对福建省2004年1月-2015年12月戊肝的月发病例数进行SARIMA分析。结果 2004年1月-2014年8月福建省戊肝月发病序列呈先升后降的趋势和周期性波动,取自然对数和1次1阶非季节差分后序列得到平稳化,模型SARIMA(0,1,1)(1,0,1)12和SARIMA(2,1,0)(1,0,1)12参数有统计学意义,残差为白噪声,后者为最优模型,表达式为:(1+0.61L+0.23L2)(1-0.89L12)(1-L)log(yt)=(1-0.89L12)εt。静态回代预测值和实际值吻合较好,相对误差取绝对值后均数为13.06%。2014年1-8月预测值的相对误差较小(5月份除外),预测标准误(S.E)较小。结论运用SARIMA模型可对戊肝发病趋势进行较准确的短期预测,可为及时、科学地研判风险提供可靠的数据基础。
[Abstract]:Objective to predict the incidence trend of hepatitis E in Fujian Province by using SARIMA model and to provide quantitative data for early warning and risk assessment. Methods the monthly incidence of hepatitis E in Fujian Province from January 2004 to December 2015 was analyzed by SARIMA with Eviews 5.0. Results from January 2004 to August 2014, the monthly incidence sequence of liver E in Fujian Province showed a trend of first rising and then decreasing, and periodic fluctuation. The natural logarithm and the first order non-seasonal difference sequence were stable. The parameters of model SARIMA0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, The residual error is white noise, the latter is the optimal model, and the expression is: 1 0. 61L 0. 23L 2U, 1-0. 89L 12, 1-L logyt (1-0. 89L12) 蔚 t. The relative error of relative error is 13.066.The relative error of forecast value in January to August of 2014 is smaller than that of May, and the error of prediction standard is smaller than that of standard error (S.E), and the static prediction value is in good agreement with the actual value, and the relative error of relative error is 13.066.The relative error of forecast value in January-August 2014 is smaller than that in May. Conclusion the SARIMA model can be used to predict the incidence trend of hepatitis E in the short term and provide a reliable data base for the timely and scientific study of the risk.
【作者单位】：福建省疾病预防控制中心福建省人兽共患病研究重点实验室;中国疾病预防控制中心应急中心监测预警与风险评估办公室;
【分类号】：R512.6;R183

【共引文献】