Modeling to COVID-19 Mortality Data: Using the Neutrosophic Gompertz Inverse Rayleigh Model with Estimation

Abstract

In this article, we introduce a new statistical model called the Neutrosophic inverse Gompertz Rayleigh (NGIR) distribution, which has excellent modelling to COVID-19 mortality data in the Netherlands. The model uses neutrosophic logic to address uncertainty in the data by representing parameters as time intervals (using neutrosophic logic, the direct method). The basic distribution functions were found, and several mathematical properties of the distribution were derived. Several tables illustrating the behavior of the distribution were developed based on these properties. Equations for estimating the parameters of the distribution were found using three estimation methods. The performance of NGIR was evaluated using Monte Carlo simulations of five estimation methods, and the results of these simulations were compared using several statistical measures to determine which method is best for estimation. Practical application on mortality data confirmed the model's ability to represent complex data with high degrees of uncertainty, making it a powerful tool for epidemiological analysis. The results demonstrated that NGIR outperforms other neutrosophic distributions in terms of information criteria and goodness of fit.