Application of Data Science for COVID‐19 Modeling in Africa: A Comparative Study of ARIMA and ANN Models, a Retrospective Time‐Series Analysis
Health Science Reports · 2026-11 · 9 (11)
DOI: 10.1002/hsr2.73498 ↗Kurz gefasst
Fachartikel in der Zeitschrift Health Science Reports, online erschienen am 10.11.2026. Verfasst von Kofi Afriyie Nyamekye und 5 weiteren Autor:innen. Beteiligte Einrichtung: University of Professional Studies Accra Ghana. Die Arbeit nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: FIC NIH HHS, Fogarty International Center.
Abstract (Originalsprache)
Background and Aims The COVID‐19 pandemic highlighted the urgent need for robust forecasting tools to guide public health responses in Africa, where data infrastructure remains limited, and epidemic dynamics differ substantially from high‐income settings.
Methods
Daily COVID‐19 surveillance data were obtained from the WHO COVID‐19 Dashboard for Egypt, Ghana, Kenya, Morocco, Nigeria, South Africa, and Tunisia. ARIMA models were fitted using the auto.arima function in R, with stationarity assessed using Augmented Dickey‐Fuller (ADF) and Kwiatkowski‐Phillips‐Schmidt‐Shin (KPSS) tests. ANN models were implemented using the nnetar algorithm, configured with lag parameter p = 7, seasonal lag p = 1, hidden layer size of 10 nodes, and 20 repeats. Model accuracy was evaluated using root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and R‐squared ( R 2 ).
Results
ARIMA outperformed ANN for four countries (Egypt, Kenya, Morocco, Nigeria), while ANN performed better for three countries (Ghana, Tunisia, South Africa). Egypt achieved the highest forecast accuracy (MAPE = 0.09% for ARIMA). South Africa presented the greatest forecasting challenge due to high volatility and multiple distinct waves, though ANN substantially outperformed ARIMA in this setting, reducing RMSE by 56% and MAPE by 51%. Negative R 2 values for six of seven countries reflected the structural inability of backward‐looking autoregressive models to anticipate variant‐driven wave peaks, consistent with limitations documented in COVID‐19 forecasting literature.
Conclusion
Neither model is universally superior; model selection should be context‐dependent. ANN is recommended for settings with stable case fatality rates and complex non‐linear dynamics (e.g., Ghana) and for high‐volume settings with multiple distinct waves (e.g., South Africa). ARIMA is recommended for moderate‐transmission settings with more linear patterns where interpretability is prioritized.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- University of Professional Studies Accra Ghana.
- University of Ghana Medical School Accra Ghana.
- University of Ghana Accra Ghana.
ORCID-Profile: Kofi Afriyie Nyamekye, Eric Sampane-Donkor, Josephine Aboagye, Haruna Latif, Justice Moses K Aheto
Förderung
- FIC NIH HHS
- Fogarty International Center
Zitieren
APA
Nyamekye, K. A., Sampane-Donkor, E., Aboagye, J., Bennin, F., Latif, H., & Aheto, J. M. K. (2026). Application of Data Science for COVID‐19 Modeling in Africa: A Comparative Study of ARIMA and ANN Models, a Retrospective Time‐Series Analysis. Health Science Reports, 9(11). https://doi.org/10.1002/hsr2.73498
Vancouver
Nyamekye KA, Sampane-Donkor E, Aboagye J, Bennin F, Latif H, Aheto JMK. Application of Data Science for COVID‐19 Modeling in Africa: A Comparative Study of ARIMA and ANN Models, a Retrospective Time‐Series Analysis. Health Science Reports. 2026;9(11). doi:10.1002/hsr2.73498
BibTeX
@article{Nyamekye2026,
author = {Nyamekye, Kofi Afriyie and Sampane-Donkor, Eric and Aboagye, Josephine and Bennin, Felicity and Latif, Haruna and Aheto, Justice Moses K},
title = {Application of Data Science for COVID‐19 Modeling in Africa: A Comparative Study of ARIMA and ANN Models, a Retrospective Time‐Series Analysis},
journal = {Health Science Reports},
year = {2026},
volume = {9},
number = {11},
doi = {10.1002/hsr2.73498}
}
EUVIMED