Time series analysis and prediction of scarlet fever incidence trends in Jiangsu Province, China: using ARIMA and TBATS models

Abstract

Abstract Objective: Scarlet fever is an increasingly serious public health problem that has attracted widespread attention worldwide. In this study, two models were constructed based on time series to predict the number of scarlet fever incidence in Jiangsu province, China Methods: Two models, ARIMA model and TBATS model, were constructed to predict the number of scarlet fever incidence in Jiangsu province, China, in the first half of 2022 based on the number of scarlet fever incidence from 2013-2021, and root mean square error (RMSE) and mean absolute percentage error (MAPE) were used to select the models and evaluate the performance of the models. Results: The incidence of scarlet fever in Jiangsu province from 2013 to 2021 was significantly bi-seasonal and trendy, and the best ARIMA model established was ARIMA(1,0,1)(2,1,1)12, with RMSE=92.23 and MAPE=47.48% for the fitting part and RMSE=138.31 and MAPE=79.11 for the prediction part. The best The best TBATS model is TBATS(0.278,{0,0}, -, {<12,5>}) with RMSE=69.85 and MAPE=27.44% for the fitted part. The RMSE of the prediction part=57.11, MAPE=39.52%. The error of TBATS is smaller than that of ARIMA model for both fitting and forecasting. Conclusion: The TBATS model outperformed the most commonly used SARIMA model in predicting the number of scarlet fever incidence in Jiangsu Province, China, and can be used as a flexible and useful tool in the decision-making process of scarlet fever prevention and control in Jiangsu Province