The Unit Gompertz Regression Model with Applications to COVID-19 Data in Brazil

In this paper, based on the unit Gompertz distribution, a new regression model for the unit interval has been introduced. In this model, the parameterization is done based on the median of the distribution. The estimation of the model parameters is done by maximum likelihood method. Monte Carlo simulations, to show the performance of maximum likelihood estimators are performed. Finally, using Covid-19 data in Brazilian capital cities, the unit Gompertz regression model performance better than the popular beta and Kumaraswamy regression models.