An Alternate to the UNH-G Family of Distributions with Illustration Download PDF

Journal Name : SunText Review of BioTechnology

DOI : 10.51737/2766-5097.2020.015

Article Type : Short commentary

Authors : Ampadu CB

Keywords : UNH-G; Weibull; Exponential; Modified; Guinea pigs data

Abstract

In this short communication we propose an alternate to the UNH-G family of distributions appearing. Further we show a sub-model of the new family is good in fitting real life data. Readers are asked to investigate further properties and applications of the new class of statistical distributions.


Contents

AMS Subject Classifications: 62exx


          Introduction to the New Family

          Practical Illustration           

          Open Problem


Introduction to the New Family

The new family can be seen as a certain modification of the Weibull or Exponential distributions, to introduce the following as its CDF

Where G is some baseline distribution with parameter vector ?, and a, b, c > 0. Obviously, the PDF can be obtained by differentiating the CDF. The new distribution is a good fit to real life data as shown in the next section [1,2].


Practical Illustration

We assume a = b = c = 1, and the baseline distribution is Dagum with the following CDF

Where x, f, g, h > 0. Thus the CDF of the new distribution (Modified Weibull-Dagum) is given by



Figure 1: The CDF and PDF of MWD(0.70896,3.48504,2.10098) fitted to the empirical distribution and histogram of the survival times of guinea pigs data.

Notation: We write  if J is a modified Weibull Dagum random variable


Open Problem

It is an open problem to obtain properties and applications of this new class of statistical distributions. The author invites readers to tackle this open problem.