[Study on the ARIMA model application to predict echinococcosis cases in China]

Area of Science:

• Epidemiology
• Biostatistics
• Infectious Disease Modeling

Background:

• Echinococcosis poses a significant public health challenge in China.
• Accurate forecasting of infectious disease trends is crucial for effective public health interventions.

Purpose of the Study:

• To develop and evaluate Autoregressive Integrated Moving Average (ARIMA) models for predicting monthly reported echinococcosis cases in China.
• To provide a data-driven reference for echinococcosis prevention and control strategies.

Main Methods:

• Time series analysis using SPSS 24.0 software.
• Construction of ARIMA models based on monthly echinococcosis case data from 2007-2015 and 2007-2014.
• Comparative analysis of the predictive accuracy of different ARIMA models.

Main Results:

• The ARIMA model utilizing data from 2007-2014 (ARIMA (1,0,0)(1,0,1)12) demonstrated higher accuracy with a relative error of 0.56%.
• The model using 2007-2015 data (ARIMA (1,0,0)(1,1,0)12) had a relative error of -13.97%.
• Statistical significance was confirmed for key model parameters (AR, SAR, SMA) in both models.

Conclusions:

• The optimal ARIMA model for predicting echinococcosis cases can vary depending on the time series data used.
• Continuous model adjustment and optimization with accumulated data are essential for dynamic prediction.
• Forecasting accuracy and prediction timeframes are influenced by data volume and reporting intensity.