Assessing Different Spatial Grouping Algorithms: An Application to the Design of Australia's New Statistical Geography

Abstract Algorithms have been used to assist in the design of statistical geographies which define areal groupings for which official data are collected. Typically the groupings based on different parameter values or different algorithms are not formally compared, but summary statistics based on each solution can be constructed, some of which are based on fuzzy set principles. These principles can also be used to make incremental adjustments to solutions, based on different criteria. In this paper, we adopt the above methodology to compare the new statistical geography for New South Wales developed by the Australian Bureau of Statistics with that generated by an amended version of the Coombes algorithm. RÉSUMÉ On a utilisé des algorithmes pour faciliter la conception de géographies statistiques définissant des groupements par zone pour lesquels des données officielles sont recueillies. En général, on ne compare pas officiellement les groupements basés sur différentes valeurs paramétriques ou différents algorithmes, mais il est possible de créer des statistiques sommaires sur la base de chaque solution, certaines desquelles sont fondées sur des ensembles de principes flous. Ces principes peuvent également être utilisés pour effectuer des modifications progressives sur des solutions, sur la base de différents critères. Dans la présente communication, nous adoptons la méthodologie susmentionnée pour comparer la nouvelle géographie statistique pour la province des Nouvelles Galles du Sud, créée par l'Australian Bureau of Statistics, avec celle qui a été produite par une version modifiée de l'algorithme de Coombes. EXTRACTO Se han utilizado algoritmos para ayudar en el diseño de geografías estadísticas que definan los agrupamientos de áreas para los que se recopilan datos oficiales. Típicamente, los agrupamientos basados en valores de parámetros diferentes o algoritmos diferentes no se comparan formalmente, pero pueden construirse estadísticas resumidas basadas en cada solución, algunas de las cuales se basan en principios establecidos difusos. Estos principios también pueden emplearse para hacer ajustes incrementales a soluciones, basados en diferentes criterios. En este estudio, adoptamos la metodología anterior para comparar la nueva geografía estadística aplicable a Nueva Gales del Sur, desarrollada por la Australian Bureau of Statistics, con la generada por una versión modificada del algoritmo de Coombes. 摘要 : 在统计地理学设计中人们引入一些算法来定义采集官方数据的不同区域 分组 。通 常情况下 ' 基于不同参数或者算法的分组方式之间不会进行正式比较 ' 但是我们可以基于每种方法构建统计概要 ' 其中有一些使用了模糊集合理论 。这些理论也可用来根据不同准则对方案进行增量调整 。本文中 ' 我们使用上述方法比较了澳大利亚统计局为新南威尔士州建立的新统计地理学和采用修订的Coombes算法 得到的统计地理学 。

[1]  Kerry L. Papps,et al.  Identifying Functional Labor Market Areas in New Zealand: A Reconnaissance Study Using Travel-to-Work Data , 2002, SSRN Electronic Journal.

[2]  C. Forster,et al.  The Challenge of Change: Australian Cities and Urban Planning in the New Millennium , 2006 .

[3]  Mike Coombes,et al.  Defining Locality Boundaries with Synthetic Data , 2000 .

[4]  L. A. Brown,et al.  THE DELIMITATION OF FUNCTIONAL REGIONS, NODAL REGIONS, AND HIERARCHIES BY FUNCTIONAL DISTANCE APPROACHES* , 1971 .

[5]  R. Stimson,et al.  Creating a New Geography of Functional Economic Regions to Analyse Aspects of Labour Market Performance in Australia , 2010 .

[6]  Zhiqiang Feng,et al.  Fuzziness of Travel-to-Work Areas , 2009 .

[7]  Ivana Paniccia,et al.  Industrial Districts: Evolution and Competitiveness in Italian Firms , 2002 .

[8]  C. Tolbert,et al.  U.S. COMMUTING ZONES AND LABOR MARKET AREAS: A 1990 UPDATE , 1996 .

[9]  James Bailey,et al.  Information Theoretic Measures for Clusterings Comparison: Variants, Properties, Normalization and Correction for Chance , 2010, J. Mach. Learn. Res..

[10]  José M. Casado-Díaz,et al.  Local Labour Market Areas in Spain: A Case Study , 2000 .

[11]  J. H. Nel,et al.  The identification of possible future provincial boundaries for South Africa based on an intramax analysis of journey-to-work data , 2008 .

[12]  Mike Coombes,et al.  Travel-to-Work Areas: the 2007 review , 2008 .

[13]  José Manuel Casado-Díaz,et al.  An evolutionary approach to the delineation of functional areas based on travel-to-work flows , 2008, Int. J. Autom. Comput..

[14]  Martin Perry,et al.  Explaining Continuity in New Zealand's Local Labour Market Areas 1991 to 2001 , 2005 .

[15]  Dion Bongaerts,et al.  Delimitation and Coherence of Functional and Administrative Regions , 2009 .

[16]  J. F. B. Goodman,et al.  THE DEFINITION AND ANALYSIS OF LOCAL LABOUR MARKETS: SOME EMPIRICAL PROBLEMS , 1970 .

[17]  Martin Watts,et al.  Rules versus Hierarchy: An Application of Fuzzy Set Theory to the Assessment of Spatial Grouping Techniques , 2009, ICANNGA.

[18]  Mike Coombes Defining Labour Market Areas by Analysing Commuting Data: Innovative Methods in the 2007 Review of Travel-to-Work Areas , 2010 .

[19]  Anne E. Green,et al.  An Efficient Algorithm to Generate Official Statistical Reporting Areas: The Case of the 1984 Travel-to-Work Areas Revision in Britain , 1986 .

[20]  Anne Kaag Andersen,et al.  Are Commuting Areas Relevant for the Delimitation of Administrative Regions in Denmark? , 2002 .

[21]  M. Coombes,et al.  The evolution of Local Labour Market Areas in contrasting region , 2005 .

[22]  Mike Coombes Travel to Work Areas and the 2001 Census , 2002 .

[23]  Stan Openshaw,et al.  Modifiable Areal Unit Problem , 2008, Encyclopedia of GIS.

[24]  Ian Masser,et al.  Hierarchical Aggregation Procedures for Interaction Data , 1975 .

[25]  Martin Watts,et al.  Local labour markets in New South Wales: fact or fiction? , 2004 .