Modeling COVID-19 Binary Data in the Aspect of Neoplasms as a Potential Indicator of Cancer by Logit and Probit Regression Models

Abstract

In this study, the effects of disability-adjusted life years (DALYs) from neoplasms and concomitant non-communicable diseases (NCDs) on total deaths from the COVID-19 pandemic until 21 July 2021 are examined globally for 179 countries. For this purpose, the explanatory variables are taken as DALYs as a measure of total burden of diseases in life lost years and lived with a disability years from neoplasm and NCDs. In this study, the total number of deaths caused by the COVID-19 pandemic has been made categorical with the help of the indicator variable and then taken as the response variable. Thus, in this study, the effects of neoplasms and concomitant NCDs on the COVID-19 pandemic are investigated by using binary logit and binary probit regression models in the family of generalized linear models (GLMs) as statistical methods. Specific to this study, the superiority of the probit model which is based on the assumption that the errors have a normal distribution in the statistical sense over the logit model which is based on the assumption that the errors have a logistic distribution is emphasized. As principle results and major conclusion from this study, neoplasms, cirrhosis and other chronic liver diseases, cardiovascular diseases, skin and subcutaneous diseases and other non-communicable diseases have been found to have statistically significant effects on deaths due to the COVID-19 pandemic.