A hot spot analysis resulting in a new layer with statistically high population values by USA county.
What is a hot spot analysis?
A hot spot analysis is the process of determining if there are statistically significant clusters in spatial data. To execute the analysis, use the spatial analysis service and the FindHotspot operation.
The analysis uses the (Gi* de Getis-Ord) equation to find groups of clusters of either point or polygon data based on high (hot) or low (cold) values, which are defined as the resultant z-scores and p-values. For a feature to be in a hot spot cluster, it must have a high value and be surrounded by other clusters with high values. For a feature to be in a cold spot cluster, the feature must have a low value and be surrounded by other clusters with low values.
A hot spot analysis helps to find spatial trends and patterns in the data that may not be visible at first glance.
Real-world examples of this analysis include the following:
Finding high and low clusters of crime or traffic crash incidents.
Discovering the distribution of high and low property values in an area.
Determining where people are spending the most money in an area compared to another.
How to perform a hot spot analysis
Review the parameters for the FindHotSpots operation.
Send a request to get the spatial analysis service URL.
Execute a job request with the following URL and parameters:
A string representing the name of the hosted feature layer to return with the results. NOTE: If you do not include this parameter, the results are returned as a feature collection (JSON).
{"serviceProperties": {"name": "<SERVICE_NAME>"}}
context
A bounding box or output spatial reference for the analysis.
"extent":{"xmin:", "ymin:", "xmax:", "ymax:"}
Code examples
Find traffic crash hot spots
This example uses the FindHotSpots operation to determine where there are statistically significant clusters of Traffic crashes counted within a fishnet grid. The red clusters represent a statistically high number of crashes. The blue spots represent a statistically low number of crashes.
In the analysis, the analysisLayer value is the Traffic crashes hosted feature layer. The points in the layer are counted within a fishnet, which was set in the shapeType parameter.
Hot spot analysis showing statistically signifcant areas of traffic crashes.
APIs
ArcGIS API for PythonArcGIS API for PythonArcGIS REST JS
This example uses the FindHotSpots operation to determine where there are statistically significant hot and cold spots for home values in Portland. The hot and cold spots, shown as red and blue respectively, indicate whether a home's value is in a statistically high or low cluster.
In the analysis, the analysisLayer value is the Enriched Portland hexagon bins hosted feature layer. The feature layer was created using generated hexagon bins that were enriched using data from the GeoEnrichment service. To analyze home values, you set the analysisField with the AVG_CY attribute.
Learn how to perform related analyses interactively with Map Viewer and programmatically with ArcGIS API for Python, ArcGIS REST JS, and ArcGIS REST API.