Comparison of Different Time Series Models in the Prediction of Hemorrhagic Fever with Renal Syndrome in Weifang

VernacularTitle:不同时间序列模型在潍坊市肾综合征出血热预测应用中的比较研究
Author: Liang ZHENG ¹ ; Qi GAO ; Shengnan YU
Author Information

1. 山东大学公共卫生学院流行病与卫生统计学系(250012)
Keywords: Hemorrhagic fever with renal syndrome; Meteorological factors; SARIMA; LSTM; EDM
From: Chinese Journal of Health Statistics 2024;41(3):393-397
CountryChina
Language:Chinese
Abstract: Objective To compare the effects of SARIMA,LSTM and EDM in predicting the incidence of HFRS in Weifang under different circumstances,and explore the best prediction model.Methods The monthly incidence of HFRS in Weifang from January 2011 to December 2017 was selected to construct the SARIMA model,univariate LSTM model,univariate EDM model,and SARIMAX model,multivariate LSTM model,and multivariate EDM model including meteorological factors.The monthly incidence from January 2018 to December 2018 was predicted,and the prediction effects of each model were compared.Results The MAPE of SARIMA model,univariate LSTM model,multivariate LSTM model,univariate EDM model,multivariate EDM model were 42.17%,48.40%,16.19%,55.00%,51.79%,respectively.Conclusion The multivariate LSTM model including meteorological factors had a good prediction effect on the incidence of HFRS in Weifang,and the prediction results could provide reference for the prevention and control of HFRS.