Colocations of spatial clusters among different industries

Inoue, R. and Shiode, Shino and Shiode, Narushige (2023) Colocations of spatial clusters among different industries. Computational Urban Science , ISSN 2730-6852.

Abstract

Spatial colocation have been studied in many contexts including locations of urban facilities, industry entities and businesses. However, identifying colocations among a small number of facilities and establishments holds the risk of introducing false positive in that such a spatial arrangement may have occurred by chance. To account for the association between a group of facilities that frequently colocate with each other, this study proposes a two-step approach consisting of identifying statistically significant clusters of each facility type using the False Discovery Rate (FDR) controlling procedure, and subsequently measuring the colocation of those clusters with the frequent-pattern-growth (FP-growth) algorithm. Empirical analysis of 6 million business and industrial establishments across Japan suggests that 10 out of 86 industry types form clear colocations and their colocations form a multi-layered, cascading structure. The number of layers in the multi-layered structure reflect the city size and the strength of the association between the colocated clusters of industries. These patterns illustrate the utility of detecting colocation of clusters towards understanding the agglomeration of different businesses. The proposed method can be applied to other contexts that would benefit from investigations into how different types of spatial features can be linked with each other and how they form colocations.

Metadata

Statistics

DownloadsShow export options

Export as XMLJSONCSV

Activity Overview

6 month trend

23Downloads

6 month trend

63Hits

Additional statistics are available via IRStats2.

Archive Staff Only (login required)

Edit/View Item