Evaluation of forecast performance of COVID-19 with different time horizons / Amirul Rashid Che Samsol and Azlan Abdul Aziz

• Main Authors: Che Samsol, Amirul Rashid, Abdul Aziz, Azlan
• 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/100729/1/100729.pdf; https://ir.uitm.edu.my/id/eprint/100729/
• Institution: Universiti Teknologi Mara

The first wave of the disease in Malaysia from 25 January to 16 February 2020 involved 22 cases. Accurate forecasting of COVID-19 case movements is crucial for the preparedness of the country’s health systems in terms of outbreak management and resource planning. The study's main goal is to generate the forecast values for COVID-19 cases in Malaysia by using forecasting models Data from the Malaysia's Ministry of Health (MOH) have been obtained from 2020 to 2022 with 1016 observations. This study aims to determine the best "win" model and produce forecast values by using Time-series Cross-Validation. Five models and three error measures have been implemented in this study. There are Naïve model, Mean Model, Single Exponential Smoothing Technique, Holt's method, and Box-Jenkins model. While the error measures used are Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) and Mean Absolute Scale Error (MASE). To execute these models, RStudio software is based on R programming language 4.2.2. The results show that the best "win" model for COVID-19 cases in Malaysia is Naïve model, Single Exponential Smoothing Technique, Holt’s Method and ARIMA(0,0,0) and mean model, respectively. The finding of this study will improve Malaysians’ decisions and awareness.