ABSTRACT
Predicting the severity and speed of transmission of the disease is crucial to resource management and developing strategies to control the spread of infectious diseases. This chapter focuses on the short-term predictions of the number of infected cases or deaths. Using various examples in COVID-19 prediction, it presents how to conduct predictive analysis in a given area via time series analysis, a well-known method in statistics for prediction. The chapter introduces basic knowledge of time series analysis, descriptive techniques, and two main popular methods for forecasting: exponential smoothing and autoregressive integrated moving average models. It uses the R packages “fable,” “feasts,” “forecast,” “tsibble” and “dplyr,” which together offer various functions for visualizing and analyzing essential time series components. Dealing with outliers is one of the earliest data analysis challenges, and since nearly all datasets contain outliers with different percentages, it continues to be one of the most critical problems to solve.
