Volume 26, Issue 125, July 2022

Simple prediction accuracy for covid-19 daily incident data in Malaysia using machine learning- Auto Regressive Integrated Moving Average (ARIMA), and Linear Regression (LR) Model

Mohamad Arif Awang Nawi^♦, Muhamamd Amirul Mat Lazin

School of Dental Sciences, Health Campus, Universiti Sains Malaysia, 16150 Kubang Kerian, Kelantan, Malaysia.

^♦Corresponding author
School of Dental Sciences, Health Campus, Universiti Sains Malaysia, 16150 Kubang Kerian, Kelantan, Malaysia

ABSTRACT

Introduction: The COVID-19 pandemic has spread so widely across borders and worldwide. All countries, including Malaysia, need to take immediate and aggressive action to curb and combat this virus by accurately predicting the number of new COVID-19 cases a few weeks in advance. This paper aims to present robust traditional methods that can be used to make accurate predictions for the evolution of an epidemic over the next two weeks. Methodology: There are two traditional methods for forecasting, such as ARIMA and LR. We use COVID-19 daily data in Malaysia from 1 January 2021 to 14 January 2022, and use Microsoft Excel, Minitab, and SPSS software version 26 to perform the forecasting procedure. Results: The ARIMA model showed a slight difference between the forecast value and the actual value of the new COVID-19 case compared to the forecast value for the LR model. This indicates that the efficient and usable prediction model is the ARIMA model with high adj-R2 values and low MAD, MSE, RMSE, and MAPE values. Conclusions: Accurately predicting trends is an important aspect of preventing the spread of COVID-19 outbreaks, especially in countries with large populations. Furthermore, accurate forecasts can provide feedback on whether the policies implemented effectively relieve the pressures of the national healthcare system and enable the government to evaluate different strategies for risk reduction and to regulate different policies based on projections of different areas of concern.

Keywords: Prediction, COVID-19, ARIMA, LR, Malaysia

Medical Science, 2022, 26, ms292e2305

PDF

DOI: https://doi.org/10.54905/disssi/v26i125/ms292e2305

Published: 20 July 2022