摘要
采用自回归移动平均模型(ARIMA)对云南省乙类传染病的月发病率进行预测研究,为传染病的防控提供参考依据。收集2005~2016年云南省年度人口数据和乙类传染病月发病率数据,针对2005~2016年云南省年度人口数据建立GM(1,1)预测模型;根据2005~2016年云南省乙类传染病月发病率数据建立ARIMA预测模型。云南省乙类传染病的发病率有很强的周期性和季节性,可通过决定系数R2、赤池信息准则(AIC)和施瓦茨准则(SC)选择出最优的乘积季节模型ARIMA(0,1,1)×(2,1,0)12来预测云南省乙类传染病的月发病率。通过对比2017年1月到10月传染病发病率的真实值和预测值,得到误差的平均值为0.8,相对误差的平均值为3.56%,说明预测效果比较满意。通过F检验和t检验显示预测值和真实值无显著性差异,说明ARIMA乘积季节模型可以较好的预测云南省乙类传染病。
An autoregressive integrated moving average(ARIMA)model was used to predict the monthly incidence of class-B infectious diseases(CBIDs)in Yunnan Province,China.In this way,we wished to provide a reference for the prevention and control of infectious diseases.Data on the annual population and monthly incidence of CBIDs of Yunnan Province,China were collected from 2005 to 2016.The GM(1,1)model was based on the annual population data of Yunnan Province from 2005 to 2016.The ARIMA model was established for analyses of the monthly incidence of CBIDs in Yunnan Province,China from 2005 to2016,and showed it to have strong periodicity and seasonality.The best product seasonal model,ARIMA(0,1,1)×(2,1,0)12,was selected through R,AIC and SC to predict CBID incidence in Yunnan Province,China.Comparison of the true and predicted incidence of CBIDs from January to October 2017 revealed the average error to be 0.8 and the relative error to be 3.56%:these data suggested that the prediction was very satisfactory.The F test and Student's t-test showed that there was no significant difference between the predicted value and true value.Hence,the ARIMA model could be used to predict the monthly incidence of CBIDs in Yunnan Province,China.
作者
李鹏
杨世宏
马磊
相艳
邵党国
韩晓东
LI Peng;YANG Shihong;MA Lei;XIANG Yan;SHAO Dangguo;HAN Xiaodong;(Faculty of Information Engineering and Automation,Kunming University of Science and Technology)
出处
《病毒学报》
CAS
CSCD
北大核心
2018年第2期201-208,共8页
Chinese Journal of Virology
基金
国家自然科学基金(项目号:61702069)
题目:癌症miRNA海绵因果网络与模块识别及其功能特性研究
云南省科技厅面上项目(项目号:KKS0201703015)
题目:分数阶微分在医学超声B模式图像中的应用研究
国家博士后科学基金(项目号:2016M592894XB)
题目:基于Web的有色金属领域主题搜索关键技术研究~~
关键词
乙类传染病
自回归移动平均模型(ARIMA)
人口数据
预测
Class-B infectious diseases
Autoregressive integrated moving average_(ARIMA) model
Pop-ulation data
Prediction