临床研究2026，Vol.34Issue(2)：13-16,4.DOI:10.12385/j.issn.2096-1278(2026)02-0013-04
自回归积分滑动平均模型和Prophet模型在河南省流行性感冒预测中的运用研究
Application of Autoregressive Integrated Moving Average and Prophet Models in Predicting Influenza in Henan Province
宋会群 1宋沛沛 2陈述3
作者信息
- 1. 开封市祥符区第二人民医院 儿科,河南 开封 475100
- 2. 开封市人民医院 急诊科,河南 开封 475100
- 3. 开封市祥符区陈留镇卫生院 公共卫生科,河南 开封 475100
- 折叠
摘要
Abstract
Objective To evaluate the predictive performance of the autoregressive integrated moving average(ARIMA)model and the Prophet model for influenza in Henan Province.and to provide evidence for influenza prevention and control in Henan Province.Methods Monthly reported cases in Henan Province from January 2016 to May 2024.published on the official website of the Henan Provincial Health Commission.were collected.A database was established using Excel 2021 to analyze temporal trends and seasonal characteristics of influenza in Henan Province.ARIMA and Prophet models were constructed using R software.Model fitting and forecasting performance were evaluated using the root mean square error(RMSE).mean absolute error(MAE).and mean absolute percentage error(MAPE).Results The peak influenza season in Henan Province was concentrated in winter and spring.Model fitting showed that the ARIMA(2,1,1)(0,0,2)12 model achieved RMSE=0.31.MAE=0.21.and MAPE=0.05.whereas the Prophet model achieved RMSE=0.43.MAE=0.32.and MAPE=0.09.For forecasting.the ARIMA(2,1,1)(0,0,2)12 model achieved RMSE=0.32.MAE=0.26.and MAPE=0.06.while the Prophet model achieved RMSE=0.68.MAE=0.57.and MAPE=0.12.Conclusion The ARIMA and Prophet models showed comparable fitting performance;however.for forecasting.the ARIMA(2,1,1)(0,0,2)12 model performed better and could more accurately predict influenza trends in Henan Province.providing a theoretical basis for scientific influenza prevention and control in Henan Province.关键词
流行性感冒/ARIMA模型/Prophet模型/模型比较Key words
influenza/ARIMA model/Prophet model/model comparison分类
医药卫生引用本文复制引用
宋会群,宋沛沛,陈述..自回归积分滑动平均模型和Prophet模型在河南省流行性感冒预测中的运用研究[J].临床研究,2026,34(2):13-16,4.
