analyzing and forecasting of coronavirus time-series data: performance comparison of machine learning and statistical models

Coronavirus is a respiratory disease caused by coronavirus 2 acute respiratory syndrome. forecasting the number of new cases and deaths can be an efficient step towards predicting costs and providing timely and sufficient facilities needed in the future. the goal of the current study is to accurately formulate and predict new cases and mortality in the future. nine prediction models are tested on the coronavirus data of yazd province as a case study. due to the evaluation criteria of root mean square error (rmse), mean square error (mse), mean absolute percentage error (mape), and mean absolute value of error (mae), the models are compared. the analysis results emphasize that, according to the mentioned evaluation criteria, the knn regression model and the bats model are the best models for predicting the cumulative cases of hospitalization of coronavirus and the cumulative cases of death, respectively. moreover, for both hospitalization and death cases, the autoregressive neural network model has the worst performance among other formulations.