Application of a data-driven XGBoost model for the prediction of COVID-19 in the USA: a time-series study

• Published in: BMJ open Vol. 12; no. 7; p. e056685
• Main Authors: Fang, Zheng-gang, Yang, Shu-qin, Lv, Cai-xia, An, Shu-yi, Wu, Wei
• Format: Journal Article
• Language: English
• Published: England British Medical Journal Publishing Group 01.07.2022; BMJ Publishing Group LTD; BMJ Publishing Group
• Series: Original research
• Subjects: China - epidemiology; Coronaviruses; COVID-19; COVID-19 - epidemiology; COVID-19 vaccines; Data science; Epidemiology; Fever; Forecasting; Humans; Incidence; Infectious diseases; Machine learning; Models, Statistical; Seasonal variations; Time series; Trends; United States - epidemiology; Variables; United States; China; United States > US; COVID-19; epidemiology
• ISSN: 2044-6055; 2044-6055
• DOI: 10.1136/bmjopen-2021-056685

ObjectiveThe COVID-19 outbreak was first reported in Wuhan, China, and has been acknowledged as a pandemic due to its rapid spread worldwide. Predicting the trend of COVID-19 is of great significance for its prevention. A comparison between the autoregressive integrated moving average (ARIMA) model and the eXtreme Gradient Boosting (XGBoost) model was conducted to determine which was more accurate for anticipating the occurrence of COVID-19 in the USA.DesignTime-series study.SettingThe USA was the setting for this study.Main outcome measuresThree accuracy metrics, mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE), were applied to evaluate the performance of the two models.ResultsIn our study, for the training set and the validation set, the MAE, RMSE and MAPE of the XGBoost model were less than those of the ARIMA model.ConclusionsThe XGBoost model can help improve prediction of COVID-19 cases in the USA over the ARIMA model.