Abstract

Local Moran and local G-statistic are commonly used to identify high-value (hot spot) and low-value (cold spot) spatial clusters for various purposes. However, these popular tools are based on the concept of spatial autocorrelation or association (SA), but do not explicitly consider if values are high or low enough to deserve attention. Resultant clusters may not include areas with extreme values that practitioners often want to identify when using these tools. Additionally, these tools are based on statistics that assume observed values or estimates are highly accurate with error levels that can be ignored or are spatially uniform. In this article, problems associated with these popular SA-based cluster detection tools were illustrated. Alternative hot spot-cold spot detection methods considering estimate error were explored. The class separability classification method was demonstrated to produce useful results. A heuristic hot spot-cold spot identification method was also proposed. Based on user-determined threshold values, areas with estimates exceeding the thresholds were treated as seeds. These seeds and neighboring areas with estimates that were not statistically different from those in the seeds at a given confidence level constituted the hot spots and cold spots. Results from the heuristic method were intuitively meaningful and practically valuable.

Highlights

  • Local Moran and local G-statistic are commonly used to identify high-value and low-value spatial clusters for various purposes

  • This paper proposes a heuristic hot spots and cold spots (HSCS) detection method which relies on user-provided threshold values to determine seeds, and these seed locations are subsequently used to determine spatial clusters, given different levels of confidence levels

  • Areas with extreme high and low estimates may not be included as part of the HSCS if their neighboring values are not highly similar, and neighbors with similar but moderate values may be identified as clusters

Read more

Summary

Methods

Chun, Y.; Griffith, D.A. Spatial Statistics and Geostatistics: Theory and Applications for Geographic Information Science and Technology; SAGE Publications Ltd.: Thousand Oaks, CA, USA, 2013. Koo, H.; Chun, Y.; Wong, D.W.S. Measuring Local Spatial Autocorrelation with Data Reliability Information. Prof. Geogr. 2021, 73, 464–480. [CrossRef]

What Are Spatial Clusters?
Current Practices of Spatial Cluster Detection
12 Rhode Island
Nature of Local SA Statistics
Nature of Statistical Estimates
Alternatives to Existing Hot Spot and Cold Spot Detection Tools
Class Separability Classification as a HSCS Detection Tool
Heuristic HSCS Identification Method
An Empirical Example
The web-based applicationimplementing implementing the the naï naïve the
Simulated Data
Heuristic detection nine
Conclusions
Findings
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call