Prediction of incidence of pulmonary tuberculosis in Zhejiang based on autoregressive integrated moving average model

  Objective   Autoregressive integrated moving average (ARIMA) model was used to predict the incidence of pulmonary tuberculosis (TB) in Zhejiang province to provides scientific basis for the precise prevention and control of pulmonary TB.

  Methods  The monthly incidence rate of pulmonary TB in Zhejiang from January 2011 to August 2021 was collected. Software R (4.0.3) was used to build the ARIMA model based on the incidence rate of TB from 2011 to 2020. The model prediction was compared with the actual data from January to August in 2021 to select some optimal models.

  Results  A total of new 374718 pulmonary TB cases were reported in Zhejiang from January 2011 to August 2021 was, showing a decrease trend. The incidence rate was relatively lower from December to February and relatively higher from March to May. The optimal model was ARIMA (2, 1, 0) (1, 1, 2)₁₂. The mean relative error (MRE) between the predicted value and the actual value of the incidence of pulmonary TB in Zhejiang from January to August in 2021 fitted by this model was 8.87%. The values of AIC, BIC, RMSE and MAPE were 95.02, 111.05, 0.30, and 4.39, respectively.

  Conclusion  The ARIMA (2, 1, 0) (1, 1, 2)₁₂ model can fit and predict the incidence trend of pulmonary TB in Zhejiang, but it needs to be adjusted dynamically according to the actual situation to improve the prediction accuracy.