Generalized Exponential Distributions
by
Rameshwar D. Gupta
University of New Brunswick, Saint John, NB, CANADA
Coauthors: Debasis Kundu

The gamma, Weibull and log-normal distributions are commonly used for analyzing any lifetime data or skewed data. Recently a new family of generalized exponential (GE) distributions was introduced and studied quite extensively by Gupta and Kundu in a series of papers. After reviewing various properties of the GE distribution, we will show that it can be used as an alternative to gamma, Weibull or log-normal distributions in many situations. The generation of GE random deviates is easy. It is possible to generate approximate gamma or normal random numbers using GE distribution and the random samples thus obtained cannot be differentiated using any statistical tests.

Date received: September 2, 2003