江西省肾综合征出血热发病率ARIMA模型及其趋势预测

发布时间：2018-06-03 01:54

本文选题：时间序列 + ARIMA模型　；参考：《现代预防医学》2017年22期

【摘要】：目的探讨自回归求和移动平均模型(Auto Regressive Integrated Moving Average,ARIMA)在江西省肾综合征出血热月发病率预测的可行性,为制定出血热防控策略提供依据。方法基于江西省2006-2015年肾综合征出血热(Hemorrhagic Fever with Renal Syndrome,HFRS)逐月发病率资料建立ARIMA模型,利用2016年各月发病率检验模型预测效果,再以2006-2016年HFRS逐月发病率构建模型预测2017年HFRS发病率。结果本研究构建的ARIMA(0,0,2)(0,1,1)12模型,拟合结果与实际发病情况基本吻合。各项参数均有统计学意义(P0.05),BIC值(Schwarz Bayesian criterion,贝叶斯信息准则)=-6.792,Ljung-Box Q=14.992,P=0.452,模型残差为白噪声;2016年各月HFRS发病率预测值与实际值动态趋势基本吻合。预测2017年江西省HFRS发病率为1.45/10万。结论 ARIMA模型能很好地模拟江西省HFRS发病率在时间序列上的变动趋势,可用于江西省HFRS发病率的短期预测研究。
[Abstract]:Objective to explore the feasibility of Auto Regressive Integrated Moving Average (ARIMA) in the prediction of hemorrhagic fever monthly rate of renal syndrome in Jiangxi Province, and to provide the basis for formulating the prevention and control strategy of hemorrhagic fever. Methods based on the 2006-2015 year's hemorrhagic fever of renal syndrome in Jiangxi province (Hemorrhagic Fever with Renal Syndrome, HFRS) The ARIMA model was established for month by month incidence data, and the results were predicted by the monthly incidence test model in 2016, and then the model of month by month incidence of 2006-2016 years HFRS was constructed to predict the incidence of HFRS in 2017. The result of this study was a model of ARIMA (0,0,2) (0,1,1) 12. The fitting results were basically consistent with the actual incidence. All the parameters had statistical significance. (P0.05), the BIC value (Schwarz Bayesian criterion, Bias information criterion) =-6.792, Ljung-Box Q=14.992, P=0.452, model residual is white noise, and the prediction value of HFRS incidence rate of each month in 2016 is basically consistent with the dynamic trend of actual value. It is predicted that the HFRS incidence in Jiangxi province in 2017 is 1.45/10 million. The trend of disease rate in time series can be used for short-term prediction of HFRS incidence in Jiangxi province.
【作者单位】：南昌大学公共卫生学院流行病与卫生统计学教研室;江西省疾病预防控制中心应急办与传染病防制所;
【基金】：江西省青年科学基金资助项目(20171BAB215051)
【分类号】：R181.3;R512.8

【相似文献】