Application of ARIMA model in predicting the incidence of hepatitis E in Yunnan Province
10.3969/j.issn.1006-2483.2025.04.008
- VernacularTitle:自回归移动平均模型在云南省戊型肝炎发病数预测中的应用
- Author:
Bilian ZHU
1
;
Yingmei TANG
1
;
Zhengrong DING
2
;
Jibo HE
2
;
Weimin BAO
3
;
Qinnian LI
1
Author Information
1. The Secend Affiliated Hospital of kunming Medical University，Kunming, Yunnan 650000 , China
2. Yunnan Provincial Center for Disease Control and Prevention , Chenggong ,Yunnan 650500 , China
3. Yunnan First People“s Hospital, Kunming , Yunnan 650000 , China
- Publication Type:Journal Article
- Keywords:
Hepatitis E;
ARIMA;
Prediction
- From:
Journal of Public Health and Preventive Medicine
2025;36(4):37-41
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the application of the Autoregressive Integrated Moving Average (ARIMA) model in predicting the number of reported hepatitis E cases in Yunnan Province，to use this model to predict the incidence trend of hepatitis E, and to provide reference for the scientific prevention and control of hepatitis E. Methods Monthly reported cases of hepatitis E in Yunnan Province from 2012 to 2021 were collected. The ARIMA model was established using SPSS 27.0, and the model was validated and parameters were optimized with data from January 2022 to December 2022. The optimal fitting model was used to predict the incidence of hepatitis E in 2023. Results Hepatitis E incidence in Yunnan Province showed a certain seasonal distribution, with most cases concentrated from March to August. All parameters of ARIMA(3,1,4)(1,1,1)12 passed statistical tests. The Ljung-Box test showed statistic Q =10.050, P = 0.346, residual sequence was a white noise sequence, and goodness-of-fit index stationary R² was 0.591. The model extrapolation effect was verified with 2022 data, and MAPE was 14.747, indicating that the model extrapolation effect was effective. The number of hepatitis E cases in Yunnan Province in 2023 was expected to be 1,086. Conclusion The ARIMA (3,1,4)(1,1,1)12 model shows good fitting performance for hepatitis E cases in Yunnan Province and can effectively predict short-term disease trends, providing a theoretical basis for formulating prevention and control measures for hepatitis E.
