基于指数平滑模型的湖北省新冠肺炎疫情预测分析

COVID-19 analysis and forecast based on Exponential Smoothing Model in Hubei Province

目的分析和预测湖北省新冠肺炎(COVID-19)疫情变化趋势。方法采用平滑指数模型对累计确诊病例数、累计治愈出院病例数、累计死亡病例数、重症病例数及危重症病例数进行拟合和预测。结果湖北省COVID-19疫情逐渐得到缓解,在2月18日进入快速“缓解期”后,3月21日进入慢速“缓解期”。采用指数平滑模型获得的拟合值与实际值的趋势基本吻合,模型拟合较好,预测结果表明在4月2日现存确诊病例数将减少至1000例以内,且主要为重症和危重症病例。结论湖北省COVID-19疫情的防控措施是有效的,指数平滑法拟合效果较好,可用于COVID-19的疫情预测。

Objective To analyze and forecast the epidemic of the new coronavirus pneumonia(COVID-19)in Hubei Province.Methods Exponential Smoothing Model was applied to fit the tendency of the number of confirmed cases,discharged cases,death cases,severe cases and critical cases.Results The epidemic of COVID-19 in Hubei Province has been gradually alleviated,the rapid remission period and slow remission period were occurred after February 18th and March 21st,respectively.The Exponential Smoothing Model was significantly fit well and the fitting values were basically consistent with the actual values.Predicted results indicated that the number of existing confirmed cases was expected to reduce to less than 1000 on April 2nd,and was mainly consist of severe and critical cases.Conclusions The prevention and control measures of COVID-19 in Hubei Province were effective,and the Exponential Smoothing Model was applicable to predicate the epidemic of COVID-19.