An Overview of Forecast Analysis with ARIMA Models during the COVID-19 Pandemic: Methodology and Case Study in Brazil

Raydonal Ospina, João A. M. Gondim, Víctor Leiva () and Cecilia Castro
Additional contact information
Raydonal Ospina: Department of Statistics, Universidade Federal de Pernambuco, Recife 50670-901, Brazil
João A. M. Gondim: Department of Mathematics, Universidade Federal de Pernambuco, Recife 50670-901, Brazil
Víctor Leiva: School of Industrial Engineering, Pontificia Universidad Católica de Valparaíso, Valparaíso 2362807, Chile
Cecilia Castro: Centre of Mathematics, Universidade do Minho, 4710-057 Braga, Portugal

Mathematics, 2023, vol. 11, issue 14, 1-18

Abstract: This comprehensive overview focuses on the issues presented by the pandemic due to COVID-19, understanding its spread and the wide-ranging effects of government-imposed restrictions. The overview examines the utility of autoregressive integrated moving average (ARIMA) models, which are often overlooked in pandemic forecasting due to perceived limitations in handling complex and dynamic scenarios. Our work applies ARIMA models to a case study using data from Recife, the capital of Pernambuco, Brazil, collected between March and September 2020. The research provides insights into the implications and adaptability of predictive methods in the context of a global pandemic. The findings highlight the ARIMA models’ strength in generating accurate short-term forecasts, crucial for an immediate response to slow down the disease’s rapid spread. Accurate and timely predictions serve as the basis for evidence-based public health strategies and interventions, greatly assisting in pandemic management. Our model selection involves an automated process optimizing parameters by using autocorrelation and partial autocorrelation plots, as well as various precise measures. The performance of the chosen ARIMA model is confirmed when comparing its forecasts with real data reported after the forecast period. The study successfully forecasts both confirmed and recovered COVID-19 cases across the preventive plan phases in Recife. However, limitations in the model’s performance are observed as forecasts extend into the future. By the end of the study period, the model’s error substantially increased, and it failed to detect the stabilization and deceleration of cases. The research highlights challenges associated with COVID-19 data in Brazil, such as under-reporting and data recording delays. Despite these limitations, the study emphasizes the potential of ARIMA models for short-term pandemic forecasting while emphasizing the need for further research to enhance long-term predictions.

Keywords: ARIMA forecasting; epidemiological forecasting; pandemic analytics; predictive modeling; public health intelligence (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.mdpi.com/2227-7390/11/14/3069/pdf (application/pdf)
https://www.mdpi.com/2227-7390/11/14/3069/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:11:y:2023:i:14:p:3069-:d:1191831

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().