基于季节性(差分整合)自回归移动平均模型的广西乙类传染病发病情况预测  

Prediction of the incidence of category B infectious diseases in Guangxi based on seasonal autoregressive integrated moving average model

作　　者：韦雪梅 杨晓祥[1,2] 韦雪芹 李娟 袁宗祥 WEI Xuemei;YANG Xiaoxiang;WEI Xueqin;LI Juan;YUAN Zongxiang(Maternity and Child Health Care of Guangxi Zhuang Autonomous Region,Nanning 530021,Guangxi,China;Guangxi Clinical Research Center for Pediatric Diseases,Nanning 530021,Guangxi,China;School of Public Health,Guangxi Medical University,Nanning 530021,Guangxi,China)

机构地区：[1]广西壮族自治区妇幼保健院,南宁市530021 [2]广西儿科疾病临床医学研究中心,南宁市530021 [3]广西医科大学公共卫生学院,南宁市530021

出　　处：《内科》2023年第3期209-214,共6页Internal Medicine

基　　金：广西儿科疾病临床医学研究中心(桂科AD22035121);广西青年科学基金项目(2021JJB140599)。

摘　　要：目的应用季节性(差分整合)自回归移动平均(SARIMA)模型预测广西乙类传染病发病情况。方法将2011年1月至2022年5月广西乙类传染病月报告发病数据作为训练集构建时间序列,拟合和构建SARIMA预测模型;以2022年6月至11月的广西乙类传染病月报告发病数据作为测试集对模型进行测试。结果广西乙类传染病的发病情况呈季节性规律,最优预测模型为SARIMA(3,1,3)(2,0,0)_(12),其预测效果平均相对误差为7.99%,预测发病例数95%CI均包含了实际发病例数。结论SARIMA(3,1,3)(2,0,0)_(12)模型能较好地拟合广西乙类传染病的发病情况,可用于疫情的短期监测。Objective To predict the incidence of category B infectious diseases in Guangxi with the seasonal autoregressive integrated moving average(SARIMA)model.Method The monthly reported incidence data of category B infectious diseases in Guangxi from January 2011 to May 2022 were used as the training set to construct a time series,and the SARIMA prediction model was fitted and constructed.The monthly reported incidence data of category B infectious diseases in Guangxi from June to November 2022 was used as the test set to test the model.Results The incidence of category B infectious diseases in Guangxi was seasonal,and the optimal prediction model was SARIMA(3,1,3)(2,0,0)_(12),with a mean relative error of the predictive validity of 7.99%,and the 95%CI of the predictive incidence included the actual incidence.Conclusion The SARIMA(3,1,3)(2,0,0)_(12) model can well fit the incidence of category B infectious diseases in Guangxi and can be used for short-term surveillance of the epidemic.

关 键 词：广西壮族自治区 乙类传染病 季节性(差分整合)自回归移动平均模型 疾病预测 

分 类 号：R181.3[医药卫生—流行病学]