ANALYSIS FACTORS AFFECTING COVID-19 MORTALITY USING COUNT REGRESSION

Abstract

The ”2019 novel coronavirus” known as “ the 2019-nCoV” or simply “COVID-19” has been declared by the World Health organization (WHO), in first quarter of 2020, as a world pandemic and a public health emergency of international concern. Alas, many details related to the COVID-19 have remained unsolved completely. The success of government strategies in fighting the COVID-19 relays mainly on the results from epidemiological or statistical studies. Statistical models play a major role in providing reliable results based on appropriate analyses. Traditional (one-part) models, mixture models and mixed-effects models for counts are used to investigate effects of the WHO-regions and Cumulated COVID-19 cases on the outcome variable COVID-19 new deaths tolls. Overall result reveals there is a strong association between number of new deaths COVID-19 with predictors including the WHO regions and cumulated cases. Besides, models that account for the overdispersion feature have smallest AICs and have reasonable regression model fits.

Keywords

• covid-19
• who
• overdispersion
• zero-inflation
• mixed-effects
• death counts