Empirical Evaluation of Alternative Time-Series Models for COVID-19 Forecasting in Saudi Arabia.

Isra Al-Turaiki,Norah Alballa,Fahad Almutlaq,Hend Alrasheed

doi:10.3390/ijerph18168660

Isra Al-Turaiki, Norah Alballa + Show 2 more

Open Access

https://doi.org/10.3390/ijerph18168660

Copy DOI

Abstract

COVID-19 is a disease-causing coronavirus strain that emerged in December 2019 that led to an ongoing global pandemic. The ability to anticipate the pandemic’s path is critical. This is important in order to determine how to combat and track its spread. COVID-19 data is an example of time-series data where several methods can be applied for forecasting. Although various time-series forecasting models are available, it is difficult to draw broad theoretical conclusions regarding their relative merits. This paper presents an empirical evaluation of several time-series models for forecasting COVID-19 cases, recoveries, and deaths in Saudi Arabia. In particular, seven forecasting models were trained using autoregressive integrated moving average, TBATS, exponential smoothing, cubic spline, simple exponential smoothing Holt, and HoltWinters. The models were built using publicly available daily data of COVID-19 during the period of 24 March 2020 to 5 April 2021 reported in Saudi Arabia. The experimental results indicate that the ARIMA model had a smaller prediction error in forecasting confirmed cases, which is consistent with results reported in the literature, while cubic spline showed better predictions for recoveries and deaths. As more data become available, a fluctuation in the forecasting-accuracy metrics was observed, possibly due to abrupt changes in the data.

Highlights

The recent COVID-19 pandemic was first identified in Wuhan, China, in December2019, and started to spread globally [1], sparking a series of responses, including countrywide lockdowns, curfews, and travel bans
Time series data collected between 1 May to 6 December 2020 were used to train long short-term memory (LSTM) and gated recurrent unit (GRU).The results show that LSTM performed best in confirmed cases in all three countries, while GRU performed best in death cases in Egypt and Kuwait
The main purpose of this work is to predict the future dynamics of COVID-19 in Saudi Arabia by applying a set of commonly used statisticalanalysis models based on historical disease data

Summary

Introduction

The recent COVID-19 pandemic was first identified in Wuhan, China, in December2019, and started to spread globally [1], sparking a series of responses, including countrywide lockdowns, curfews, and travel bans. The recent COVID-19 pandemic was first identified in Wuhan, China, in December. The most common symptoms of COVID-19 infection are mild, it may have serious and even fatal effects on some patients. COVID-19 is a global crisis, with globally more than 179,686,071 confirmed cases and more than 3,899,172 deaths as of 25 June 2021 [2]. The rising number of COVID-19 cases has globally overburdened healthcare facilities, but the virus continues to be poorly understood. Researchers from different fields have been researching the COVID-19 virus since its first appearance. Predicting the progress of COVID-19 is crucial for public-health planning and decision making. One way to achieve this is by accurately estimating the number of active cases at any given point in time

Objectives

Methods

Results

Discussion

Conclusion