|
Activity Number:
|
538
- SPEED: Predictive Analytics with Social/Behavioral Science Applications: Spatial Modeling, Education Assessment, Population Behavior, and the Use of Multiple Data Sources
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2018 : 10:30 AM to 11:15 AM
|
|
Sponsor:
|
Social Statistics Section
|
|
Abstract #332542
|
|
|
Title:
|
Spatial Proximity Between Bank Branch Closures and Openings: Where Are the New Underserved Banking Areas Located?
|
|
Author(s):
|
Anna Tranfaglia*
|
|
Companies:
|
|
|
Keywords:
|
Branch Closures;
Spatial Clustering;
GIS;
Point Process
|
|
Abstract:
|
As more consumers take advantage of online banking services, branch networks are condensing across the country. Limited attention has been given to identifying spatial patterns of branch closures and openings during this era of branch network transition. Using a unique data set of complete branch bank addresses and their corresponding opening/closing dates, this paper locates and maps spatial cluster activity of diminishing (and increasing) physical access to banking services. Local K-function estimation is done to examine spatial local clustering of both new branch openings and closures from 2010 to 2016 in three metropolitan statistical areas (MSAs). By utilizing a local K-function rather than the global spatial statistic, the resulting p-values for each point i within the point process can be mapped. This results in both an empirical and visual check for clustering within each MSA studied. Preliminary global cluster analysis indicates that clustering of branch closures and openings are significant at short distances (2-4 miles) within MSAs. However, local cluster analysis is needed to fully identify the spatial structure of bank branch networks.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
|
