Will COVID-19 confirmed cases in the USA reach 3 million? A forecasting approach by using SutteARIMA Method

Objectives: Forecasting the number of COVID-19 cases in the USA can provide an overview and projection of the development of COVID-19 cases in the US so that policy makers can determine the steps that must be taken. This study aimed to determine whether COVID-19 confirmed cases in the USA would reac...

Full description

Bibliographic Details
Main Authors: Ansari Saleh Ahmar, Eva Boj
Format: Article
Language:English
Published: Elsevier 2020-11-01
Series:Current Research in Behavioral Sciences
Subjects:
USA
Online Access:http://www.sciencedirect.com/science/article/pii/S2666518220300024
Description
Summary:Objectives: Forecasting the number of COVID-19 cases in the USA can provide an overview and projection of the development of COVID-19 cases in the US so that policy makers can determine the steps that must be taken. This study aimed to determine whether COVID-19 confirmed cases in the USA would reach 3 million cases with the SutteARIMA method forecasting approach. Methods: Data from this study were obtained from the Worldometer data on 15 February 2020 to 2 July 2020. Data from 15 February 2020 to 25 June 2020 were used to performance data fitting (26 June 2020–2 July 2020). Data fitting is used to examine the extent of the accuracy of the SutteARIMA method in predicting data. To examine the level of the data accuracy, the MAPE method was used in this study. Results: The results of forecasting data fitting on 26 June 2020 – 2 July 2020: 2,544,732; 2,590,888; 2,632,477; 2,671,055; 2,711,798; 2,755,128; 2,803,729. The accuracy of SutteARIMA for the period of 26 June 2020–2 July 2020 based on MAPE was 0.539% and the forecasting results that had been obtained were 3 million confirmed cases, namely from 05 to 06 June 2020: 1,981,299; 2,005,706; 2,030,283; 2,055,031. Conclusions: The SutteARIMA method predicted that 2 million confirmed cases of COVID-19 will be obtained on the WHO situation report on days 168–170 or 05–07 June 2020.
ISSN:2666-5182