Comparison between ARIMA model and fuzzy time series: forecasting endemic COVID-19 cases in Malaysia / Nur Atikah Mohd Razali and Nor Azriani Mohamad Nor

• Main Authors: Mohd Razali, Nur Atikah, Mohamad Nor, Nor Azriani
• Format: Book Section
• Language: English
• Published: College of Computing, Informatics and Media, UiTM Perlis 2023
• Subjects: Time-series analysis
• Online Access: https://ir.uitm.edu.my/id/eprint/100269/1/100269.pdf; https://ir.uitm.edu.my/id/eprint/100269/
• Institution: Universiti Teknologi Mara

The coronavirus disease 2019 (COVID-19) began in December 2019, with Wuhan, China serving as the originating of the disease. Chinese government disclosed the discovery of the new coronavirus to the world, and the World Health Organization (WHO) declared the confirmation of the virus. Up to January 25, 2020, the virus began to spread in Malaysia from the three Chinese nationals who had previously had intimate contact with an infected individual in Singapore. On April 1, 2022, Malaysia announced the transition phase of COVID-19 from pandemic to endemic due to the success of COVID-19 vaccination. However, despite the endemic phase, the cases of COVID-19 still persist in Malaysia and surpassed thousands of cases daily. Thus, this led to this study on forecasting the endemic COVID-19 cases in Malaysia by comparing two methods to find the best model. In order to meet the aim, Fuzzy Time Series (FTS) Chen Model and ARIMA method were used to evaluate the endemic cases of COVID-19. The best model was chosen by evaluating the smallest value of Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE). ARIMA (3,1,2) model was chosen as the best model to forecast the endemic cases of COVID-19 since it generated smallest value of error measure.