Answers are grounded in this paper and cite their sources.
Statistical distributions are the foundations of statistical methodology in both theory and applications. They are the back bone of every parametric statistical method, including inference, modeling, survival, reliability, and others. In recent years, partly due to the advanced computing technology, there have been a series of developments of new methodology for generating new families of statistical distributions, which have greatly enhanced parametric statistical methods for handling real world scenarios that could not be modeled using existing distributions. One main reason for the need of generalized families is that each of the useful basic statistical distributions has its own weakness in real-world applications. The real-world phenomena are often much more complex for these commonly known basic statistical distributions to provide adequate fit. For example, Johnson et al. ( 2005 ) presented various modifications and generalizations of the Poisson distribution. Some of these distributions were developed in an attempt to explain the unequal mean and variance in the numerical data observed in different fields of applications.
No takes yet. Share an insight, caveat, or question.
Famoye et al. (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: