Rank-based Markov chains for regional income distribution dynamics

Sergio Rey

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Markov chains have become a mainstay in the literature on regional income distribution dynamics and convergence. Despite its growing popularity, the Markov framework has some restrictive characteristics associated with the underlying discretization income distributions. This paper introduces several new approaches designed to mitigate some of the issues arising from discretization. Based on the examination of rank distributions, two new Markov-based chains are developed. The first explores the movement of individual economies through the income rank distribution over time. The second provides insight on the movements of ranks over geographical space and time. These also serve as the foundation for two new tests of spatial dynamics or the extent to which movements in the rank distribution are spatially clustered. An illustration of these new methods is included using income data for the lower 48 US states for the years 1929-2009.

Original languageEnglish (US)
Pages (from-to)115-137
Number of pages23
JournalJournal of Geographical Systems
Volume16
Issue number2
DOIs
StatePublished - 2014

Fingerprint

income distribution
Markov chain
income
popularity
examination
economy
distribution
time

Keywords

  • Convergence
  • Markov
  • Regional income distributions
  • Spatial dynamics

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Earth-Surface Processes

Cite this

Rank-based Markov chains for regional income distribution dynamics. / Rey, Sergio.

In: Journal of Geographical Systems, Vol. 16, No. 2, 2014, p. 115-137.

Research output: Contribution to journalArticle

@article{4d15d1070da44a1e9238926682d044d1,
title = "Rank-based Markov chains for regional income distribution dynamics",
abstract = "Markov chains have become a mainstay in the literature on regional income distribution dynamics and convergence. Despite its growing popularity, the Markov framework has some restrictive characteristics associated with the underlying discretization income distributions. This paper introduces several new approaches designed to mitigate some of the issues arising from discretization. Based on the examination of rank distributions, two new Markov-based chains are developed. The first explores the movement of individual economies through the income rank distribution over time. The second provides insight on the movements of ranks over geographical space and time. These also serve as the foundation for two new tests of spatial dynamics or the extent to which movements in the rank distribution are spatially clustered. An illustration of these new methods is included using income data for the lower 48 US states for the years 1929-2009.",
keywords = "Convergence, Markov, Regional income distributions, Spatial dynamics",
author = "Sergio Rey",
year = "2014",
doi = "10.1007/s10109-013-0189-0",
language = "English (US)",
volume = "16",
pages = "115--137",
journal = "Journal of Geographical Systems",
issn = "1435-5930",
publisher = "Springer Verlag",
number = "2",

}

TY - JOUR

T1 - Rank-based Markov chains for regional income distribution dynamics

AU - Rey, Sergio

PY - 2014

Y1 - 2014

N2 - Markov chains have become a mainstay in the literature on regional income distribution dynamics and convergence. Despite its growing popularity, the Markov framework has some restrictive characteristics associated with the underlying discretization income distributions. This paper introduces several new approaches designed to mitigate some of the issues arising from discretization. Based on the examination of rank distributions, two new Markov-based chains are developed. The first explores the movement of individual economies through the income rank distribution over time. The second provides insight on the movements of ranks over geographical space and time. These also serve as the foundation for two new tests of spatial dynamics or the extent to which movements in the rank distribution are spatially clustered. An illustration of these new methods is included using income data for the lower 48 US states for the years 1929-2009.

AB - Markov chains have become a mainstay in the literature on regional income distribution dynamics and convergence. Despite its growing popularity, the Markov framework has some restrictive characteristics associated with the underlying discretization income distributions. This paper introduces several new approaches designed to mitigate some of the issues arising from discretization. Based on the examination of rank distributions, two new Markov-based chains are developed. The first explores the movement of individual economies through the income rank distribution over time. The second provides insight on the movements of ranks over geographical space and time. These also serve as the foundation for two new tests of spatial dynamics or the extent to which movements in the rank distribution are spatially clustered. An illustration of these new methods is included using income data for the lower 48 US states for the years 1929-2009.

KW - Convergence

KW - Markov

KW - Regional income distributions

KW - Spatial dynamics

UR - http://www.scopus.com/inward/record.url?scp=84897089001&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897089001&partnerID=8YFLogxK

U2 - 10.1007/s10109-013-0189-0

DO - 10.1007/s10109-013-0189-0

M3 - Article

AN - SCOPUS:84897089001

VL - 16

SP - 115

EP - 137

JO - Journal of Geographical Systems

JF - Journal of Geographical Systems

SN - 1435-5930

IS - 2

ER -