The Dual-Dagum family of distributions: Properties, regression and applications to COVID-19 data

Abstract

A new Dual-Dagum-G (DDa-G) family is defined as a good competitor to the Beta-G and Kumaraswamy-G generators, which are widely applied in several areas. Some of its mathematical properties are addressed. We obtain the maximum likelihood estimates, and some simulations prove the consistency of the estimates. The flexibility of this family is shown through a COVID-19 data set. We propose a new regression based on a special distribution of the DDa-G family, and provide a sensitivity analysis by using data from 1,951 COVID-19 patients collected in Curitiba, Brazil.