Paper Summary
Share...

Direct link:

Logistic-Quantile and Uniform-Based Mixture Distribution With a Doubling Technique

Thu, May 4, 8:00 to 9:30am CDT (8:00 to 9:30am CDT), Division D Virtual Sessions, Division D - Section 2: Quantitative Methods and Statistical Theory Virtual Paper Room

Abstract

This paper introduces a family of probability distributions by mixing the quantile function of standard logistic distribution and its cumulative distribution function (cdf) using a doubling technique. Specifically, this proposed family of distributions is constructed using a piecewise function of the cubic term of the quantile function of standard logistic distribution and its cdf represented by a uniform (0, 1) random variate. Also presented is a characterization of this family of distributions through the method of L-moments and conventional moments. Examples of fitting real world educational data are also presented. One of the advantages of this family of distributions over other existing families of non-normal distributions is that it can fit data that have negative values of skewness and/or kurtosis.

Author