Modeling Nigerian Covid-19 cases: A comparative analysis of models and estimators

•COVID-19 remains a major pandemic currently facing all the countries of the world, Nigeria inclusive.•We build some statistical model to the available data.•We choose the best model to make some forecast. COVID-19 remains a major pandemic currently threatening all the countries of the world. In Nig...

Full description

Saved in:

Bibliographic Details
Published in:	Chaos, solitons and fractals Vol. 138; p. 109911
Main Authors:	Ayinde, Kayode, Lukman, Adewale F., Rauf, Rauf I., Alabi, Olusegun O., Okon, Charles E., Ayinde, Opeyemi E.
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 01-09-2020
Subjects:	COVID-19 Curve Estimation Statistical Models Estimators Forecast Values Quartic Linear Regression Model COVID-19 Quartic Linear Regression Model Forecast Values Curve Estimation Statistical Models Estimators
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	•COVID-19 remains a major pandemic currently facing all the countries of the world, Nigeria inclusive.•We build some statistical model to the available data.•We choose the best model to make some forecast. COVID-19 remains a major pandemic currently threatening all the countries of the world. In Nigeria, there were 1, 932 COVID-19 confirmed cases, 319 discharged cases and 58 deaths as of 30th April 2020. This paper, therefore, subjected the daily cumulative reported COVID-19 cases of these three variables to nine (9) curve estimation statistical models in simple, quadratic, cubic, and quartic forms. It further identified the best of the thirty-six (36) models and used the same for prediction and forecasting purposes. The data collected by the Nigeria Centre for Disease Control for sixty-four (64) days, two (2) months and three (3), were daily monitored and eventually analyzed. We identified the best models to be Quartic Linear Regression Model with an autocorrelated error of order 1 (AR(1)); and found the Ordinary Least Squares, Cochrane Orcutt, Hildreth–Lu, and Prais-Winsten and Least Absolute Deviation (LAD) estimators useful to estimate the models’ parameters. Consequently, we recommended the daily cumulative forecast values of the LAD estimator for May and June 2020 with a 99% confidence level. The forecast values are alarming, and so, the Nigerian Government needs to hastily review her activities and interventions towards COVID-19 to provide some tactical and robust structures and measures to avert these challenges.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	0960-0779 1873-2887 0960-0779
DOI:	10.1016/j.chaos.2020.109911