Analysis and Prediction of COVID-19 Data in Taiwan

Abstract

Started in 2019, every country has been affected by COVID-19 pandemic, and the number of infected cases and deaths is still increasing. When other countries have a larger number of infected cases and steeper increasing trends, Taiwan has less number of infected cases and the trend of infected cases is quite stable because of early preparation. To better understand the growth of the number of infected cases in Taiwan, we aim to construct predictive models to analyze the historical data and also predict the future number of infected cases. To build up predictive models and study predictions, we examine some methods in time series analysis. Through comprehensive analysis, we find that the predicted number of infected cases is still stable and a strictly increasing trend seems not to exist.

Note: Funding: No funding to support this research.

Declaration of Interest: No conflict of interests.

Keywords: COVID-19; infection; modeling; neural network; prediction; time-dependent; time series analysis