摘要
目的:运用季节性自回归移动平均混合(SARIMA)模型分析预测河北省手足口病发病趋势的可行性和适用性,为手足口病的防控工作提供决策依据。方法:利用R3.5.1软件对河北省2008年1月至2014年12月手足口病月发病率资料进行建模,并以2015年手足口病月发病率资料验证模型的预测效果。结果:模型SARIMA(1,1,1)(0,1,1)12较好地拟合了河北省手足口病月发病率资料,模型残差为白噪声序列(Ljung-Box Q=9.289,P=0.411),预测值与实际值的相对误差范围为0.924%~35.526%,平均相对误差为13.408%。结论:SARIMA模型可较好地反映河北省手足口病的发病趋势并进行短期预测。
Objective:To discuss the feasibility and applicability of multiple seasonal autoregressive integrated moving average(SARIMA)model on forecasting incidence trend of hand-foot-mouth disease(HFMD)in Hebei province,and to provide decision basis for preventing and controlling the disease.Methods:The model was established based on monthly HFMD incidence data from Jan.2008 to Dec.2014,and monthly HFMD incidences in the year of 2015 were predicted with the established model.The R3.5.1 software was applied in the analyses.Results:SARIMA(1,1,1)(0,1,1)12 fitted well with the monthly HFMD incidence in Hebei province,and the residuals of the model were white noise(Ljung-Box Q=9.289,P=0.411).The relative error range between predicted and actual values was 0.924%-35.526%,and the average relative error was 13.408%.Conclusion:SARIMA model can well reflect the trend of HFMD and predict the incidence of HFMD in a short term in Hebei province.
作者
韩玲
王鸿
颜隆
高治理
贺娟
HAN Ling;WANG Hong;YAN Long;GAO Zhi-li;HE Juan(School of Traditional Chinese Medicine,Beijing University of Chinese Medicine,Beijing 100029,China)
出处
《中华中医药杂志》
CAS
CSCD
北大核心
2019年第12期5904-5907,共4页
China Journal of Traditional Chinese Medicine and Pharmacy
基金
国家自然科学基金项目(No.81574098).
