Describing the Flexibility of the Generalized Gamma and Related Distributions

Matthew Matheson; Alvaro Muñoz; Christopher Cox

doi:10.1186/s40488-017-0072-5

Describing the Flexibility of the Generalized Gamma and Related Distributions

Matthew Matheson, Alvaro Muñoz, Christopher Cox

Bloomberg School of Public Health

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

The generalized gamma (GG) distribution is a widely used, flexible tool for parametric survival analysis. Many alternatives and extensions to this family have been proposed. This paper characterizes the flexibility of the GG by the quartile ratio relationship, log(Q2/Q1)/log(Q3/Q2), and compares the GG on this basis with two other three-parameter distributions and four parent distributions of four or five parameters. For most parameter combinations of other distributions, a very similar GG, as assessed by the Kullback-Liebler distance, can be found by matching the three quartiles; extreme cases where this fails are examined. Limited additional flexibility is observed, supporting the basic GG family as an ideal platform for parametric survival analysis.

Original language	English (US)
Article number	15
Journal	Journal of Statistical Distributions and Applications
Volume	4
Issue number	1
DOIs	https://doi.org/10.1186/s40488-017-0072-5
State	Published - Dec 1 2017

Keywords

Generalized gamma
Hazard functions
Kullback-Liebler distance
Parametric survival analysis

ASJC Scopus subject areas

Statistics and Probability
Computer Science Applications
Statistics, Probability and Uncertainty

Access to Document

10.1186/s40488-017-0072-5

Cite this

@article{ea848f70ddf84b25a67c483f94363b43,

title = "Describing the Flexibility of the Generalized Gamma and Related Distributions",

abstract = "The generalized gamma (GG) distribution is a widely used, flexible tool for parametric survival analysis. Many alternatives and extensions to this family have been proposed. This paper characterizes the flexibility of the GG by the quartile ratio relationship, log(Q2/Q1)/log(Q3/Q2), and compares the GG on this basis with two other three-parameter distributions and four parent distributions of four or five parameters. For most parameter combinations of other distributions, a very similar GG, as assessed by the Kullback-Liebler distance, can be found by matching the three quartiles; extreme cases where this fails are examined. Limited additional flexibility is observed, supporting the basic GG family as an ideal platform for parametric survival analysis.",

keywords = "Generalized gamma, Hazard functions, Kullback-Liebler distance, Parametric survival analysis",

author = "Matthew Matheson and Alvaro Mu{\~n}oz and Christopher Cox",

note = "Publisher Copyright: {\textcopyright} 2017, The Author(s).",

year = "2017",

month = dec,

day = "1",

doi = "10.1186/s40488-017-0072-5",

language = "English (US)",

volume = "4",

journal = "Journal of Statistical Distributions and Applications",

issn = "2195-5832",

publisher = "Springer Open",

number = "1",

}

TY - JOUR

T1 - Describing the Flexibility of the Generalized Gamma and Related Distributions

AU - Matheson, Matthew

AU - Muñoz, Alvaro

AU - Cox, Christopher

PY - 2017/12/1

Y1 - 2017/12/1

N2 - The generalized gamma (GG) distribution is a widely used, flexible tool for parametric survival analysis. Many alternatives and extensions to this family have been proposed. This paper characterizes the flexibility of the GG by the quartile ratio relationship, log(Q2/Q1)/log(Q3/Q2), and compares the GG on this basis with two other three-parameter distributions and four parent distributions of four or five parameters. For most parameter combinations of other distributions, a very similar GG, as assessed by the Kullback-Liebler distance, can be found by matching the three quartiles; extreme cases where this fails are examined. Limited additional flexibility is observed, supporting the basic GG family as an ideal platform for parametric survival analysis.

AB - The generalized gamma (GG) distribution is a widely used, flexible tool for parametric survival analysis. Many alternatives and extensions to this family have been proposed. This paper characterizes the flexibility of the GG by the quartile ratio relationship, log(Q2/Q1)/log(Q3/Q2), and compares the GG on this basis with two other three-parameter distributions and four parent distributions of four or five parameters. For most parameter combinations of other distributions, a very similar GG, as assessed by the Kullback-Liebler distance, can be found by matching the three quartiles; extreme cases where this fails are examined. Limited additional flexibility is observed, supporting the basic GG family as an ideal platform for parametric survival analysis.

KW - Generalized gamma

KW - Hazard functions

KW - Kullback-Liebler distance

KW - Parametric survival analysis

UR - http://www.scopus.com/inward/record.url?scp=85065129621&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065129621&partnerID=8YFLogxK

U2 - 10.1186/s40488-017-0072-5

DO - 10.1186/s40488-017-0072-5

M3 - Article

AN - SCOPUS:85065129621

SN - 2195-5832

VL - 4

JO - Journal of Statistical Distributions and Applications

JF - Journal of Statistical Distributions and Applications

IS - 1

M1 - 15

ER -

Describing the Flexibility of the Generalized Gamma and Related Distributions

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this