首页 | 本学科首页   官方微博 | 高级检索  
     

时间序列模型在广州市登革热发病趋势预测的应用
引用本文:方钦,王颖,刘仲明,王锟,郭钜旋. 时间序列模型在广州市登革热发病趋势预测的应用[J]. 江苏预防医学, 2020, 0(1): 23-25
作者姓名:方钦  王颖  刘仲明  王锟  郭钜旋
作者单位:;1.广州市海珠区疾病预防控制中心;2.对外经济贸易大学保险学院
基金项目:2017年度海珠区区属基层医疗卫生专项项目(海科工商计2018-33)。
摘    要:目的利用ARIMA模型对广州市登革热发病趋势进行时间序列分析和预测。方法收集广州市2015-2018年登革热每周发病数,采用2015-2017年登革热每周发病数构建ARIMA模型,通过2018年登革热发病预测值与实际值的拟合情况,评价模型预测效果。结果拟合模型ARIMA(4,1,9),残差序列为白噪声,预测值与实际值平均绝对误差为4.03,均方根误差为8.13。2018年登革热预测发病趋势与实际发病趋势较吻合。结论 ARIMA模型能较好地模拟广州市登革热的短期发病趋势,可作为预测工具。

关 键 词:登革热  时间序列分析  ARIMA模型  预测

Application of time series model in predicting the incidence of dengue fever prevalence trend in Guangzhou
FANG Qin,WANG Ying,LIU Zhong-ming,WANG Kun,GUO Ju-xuan. Application of time series model in predicting the incidence of dengue fever prevalence trend in Guangzhou[J]. Jiangsu Journal of Preventive Medicine, 2020, 0(1): 23-25
Authors:FANG Qin  WANG Ying  LIU Zhong-ming  WANG Kun  GUO Ju-xuan
Affiliation:(Guangzhou Haizhu District Center for Disease Control and Prevention,GuangDong GuangZhou 510288,China;不详)
Abstract:Objective To analyze and predict incidence of dengue fever(DF)in Guangzhou with an auto regressive integrated moving average(ARIMA)model.Methods DF weekly incidence was collected from 2015 to 2018,and a ARIMA model was fit using data from 2015 to 2017.The prediction performance was evaluated by comparison of reported and predicted DF incidences in 2018.Results The model ARIMA(4,1,9)was established with residual sequence of a white noise sequence.The average absolute error between the predicted and actual values was 4.03,and the root mean square error was 8.13.The predicted epidemic trend in 2018 was consistent with the reported data.Conclusions The established ARIMA model can predict the shot-time trend of DF in Guangzhou with good performance.It can be used as an applicable tool for prediction.
Keywords:Dengue fever  Time series analysis  ARIMA model  Prediction
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号