应用ARIMA-GRNN模型对肾综合征出血热发病率时间序列数据的预测研究

发布时间：2018-01-26 19:21

本文关键词： 肾综合征出血热时间序列发病率预测　出处：《中国卫生统计》2015年02期 　论文类型：期刊论文

【摘要】：目的阐述ARIMA-GRNN模型预测肾综合征出血热发病率的方法和步骤,探讨其在综合征出血热发病率预测中的应用。方法利用辽宁省1962-2008年的肾综合征出血热发病率时间序列数据作为训练集,建立ARIMA模型和ARIMA-GRNN模型,选取2009-2011年的数据作为检验集,评价模型的拟合和预测效果。结果 ARIMA(2,1,1)模型和ARIMA-GRNN模型拟合值的平均误差绝对值分别为1.14和0.77;预测值的平均误差绝对值分别为0.53和0.20。ARIMA-GRNN模型的拟合和预测效果均优于ARIMA模型。结论 ARIMA-GRNN模型能有效模拟、预测肾综合征出血热的发病疫情,具有较强的推广应用价值。
[Abstract]:Objective to describe the method and procedure of predicting the incidence of hemorrhagic fever with renal syndrome (HFRS) by ARIMA-GRNN model. Methods the time series data of incidence rate of hemorrhagic fever with renal syndrome (HFRS) from 1962 to 2008 in Liaoning Province were used as training set. The ARIMA model and the ARIMA-GRNN model were established, and the data from 2009-2011 were selected as the test set to evaluate the fitting and prediction effects of the model. 1) the average absolute error of the fitting values of the model and the ARIMA-GRNN model is 1.14 and 0.77, respectively. The average error absolute values of the predicted values are 0.53 and 0.20.ARIMA-GRNN, respectively. The fitting and prediction results of the model are better than that of the ARIMA model. ARIMA-GRNN model can be effectively simulated. The prediction of hemorrhagic fever with renal syndrome (HFRS) is of great value in popularization and application.
【作者单位】：中国医科大学公共卫生学院;辽宁省疾病预防控制中心;
【基金】：国家自然科学基金项目(No.81202254);国家自然科学基金项目(No.30771860)
【分类号】：R512.8
【正文快照】： 肾综合征出血热(hemorrhagic fever with renalsyndrome,HFRS)是由布尼亚病毒科的汉坦病毒属引起,包括汉坦病毒、汉城病毒、obrava-Belgrade病毒和普马拉病毒[1]。HFRS流行范围广、病死率高,已经严重危害人民的生命健康,属于我国重点防治传染病之一。控制HFRS有效措施之一就是

【参考文献】