Hello to all GeoNet members!
I am in working for my final project for crime hotspot analysis.
I have crime a data that looking like clustering, so i wonder which hotspot analysis suitable for my tutorial?
Can someone here tell me on how to get the exact value of intensity for each crime if I want to using
spatial statistic tool (Getis-Ord G) ?
Need to run some tools to get that intensity value? or just randomly state the value based the weightage of that crime cases?
Thanks in advance!
It sounds like your data is crime incident points. Is this correct? The tool that you mentioned, Hot Spot Analysis (Getis-Ord Gi*), needs a variety of values for whatever variable you are trying to analyze. So if you had police beats or census tracts with the number of crimes committed in each one, you could use Hot Spot Analysis (Getis-Ord Gi*) with beats/tracts as your input and number of crimes as your variable.
If your point dataset doesn’t have an existing “intensity value” or count variable that you would like to analyze, you can either aggregate your points or use the Optimized Hot Spot Analysis tool.
To aggregate your crime incident points, use the Integrate tool to snap points within a certain distance together and then use the Collect Events tool to generate a new feature class that creates one point for each unique location and contains the number of events/snapped points as an attribute field. You could then use that number of snapped points for your input into Hot Spot Analysis (Getis-Ord Gi*). This is just one way to aggregate your points into groups…
The Optimized Hot Spot Analysis tool pretty much does all of this for you! It automatically aggregates your data and identifies the proper scale of analysis. It is important to understand the settings that this tool found to be optimal so be sure to read and understand the results in the Results window. Here is more about how this tool works: http://resources.arcgis.com/en/help/main/10.2/index.html#//005p00000058000000