Hot spot analysis - how to set distance

[HuaqiYuan]

I have a question about conducting hot spot analysis. Two criteria needs to satisfy for setting up fixed distance band.  Guaranteeing at least one neighbor and also having the peak z value.  But I found out the there is conflict to satisfy both conditions at same time.  That is, the distance to guarantee at least one neighbor sometimes is not the distance for getting the peak z value.  And the distance for getting the peak z score is not the distance for guaranteeing at least one neighbor.  I am wondering under this condition, which distance I should use for the fixed distance band.  The one guaranteeing one neighbor or the one having the peak z value.

Thanks in advance for your help.

[GavinJackson1]

I've recently done some hotspot analysis and i used the maximum distance for each point to have at least one neighbour as the deyermining factor because a large number of points don't have a neighbour, they are discounted from the hotspot analysis.

I would suggest viewing the spatial statistics video here for more info:

http://training.esri.com/acb2000/showdetl.cfm?DID=6&Product_ID=1002

Cheers

[HuaqiYuan]

I've recently done some hotspot analysis and i used the maximum distance for each point to have at least one neighbour as the deyermining factor because a large number of points don't have a neighbour, they are discounted from the hotspot analysis.

I would suggest viewing the spatial statistics video here for more info:

http://training.esri.com/acb2000/showdetl.cfm?DID=6&Product_ID=1002

Cheers

Thanks!  Is it going to be a problem if the maximum distance is not the distance where the peak Z value has.  In general, the maximum distance for having at least 1 neighbor is greater than the distance for having the peak Z value.

[LaurenRosenshein]

Hi Franky,

I'm glad to hear you've found the resources and are looking at the peak z-score distances to choose a distance band for hot spot analysis.  Hopefully you're using the Incremental Spatial Autocorrelation tool to do this, which you can find in our Supplementary Spatial Statistics toolbox.  This is a great question that you bring up, and its especially common, for instance, when you have a lot of features with similar sizes (small polygons in the more urban areas), and then a couple of features that are much larger (big polygons in more rural/suburban areas). 

A good option for dealing with this is to create a Spatial Weights Matrix.  From the Generate Spatial Weights Matrix tool, you can set the Distance Band to the distance at which the z-score peaks, and then use the Number of Neighbors parameter to make sure that for those features that don't have any neighbors at the distance that you chose, they'll have at least the number of neighbors that you set.  So, for example, if you choose a distance of 500m, and there are a couple of features that don't have any neighbors at 500m, but you also set a Number of Neighbors at 2...then for those features WITHOUT any neighbors at the specified distance band we'll increase the threshold to ensure those particular features have at least 2 neighbors.  You can only do it using the Generate Spatial Weights Matrix option, and then from Hot Spot Analysis you'll choose to Get Spatial Weights From File.  We often use this option when dealing with this issue.

Hope this helps!

Lauren Rosenshein
Geoprocessing Product Engineer

Check out the latest Spatial Statistics resources at http://esriurl.com/spatialstats

[HuaqiYuan]

Hi Franky,

I'm glad to hear you've found the resources and are looking at the peak z-score distances to choose a distance band for hot spot analysis.  Hopefully you're using the Incremental Spatial Autocorrelation tool to do this, which you can find in our Supplementary Spatial Statistics toolbox.  This is a great question that you bring up, and its especially common, for instance, when you have a lot of features with similar sizes (small polygons in the more urban areas), and then a couple of features that are much larger (big polygons in more rural/suburban areas). 

A good option for dealing with this is to create a Spatial Weights Matrix.  From the Generate Spatial Weights Matrix tool, you can set the Distance Band to the distance at which the z-score peaks, and then use the Number of Neighbors parameter to make sure that for those features that don't have any neighbors at the distance that you chose, they'll have at least the number of neighbors that you set.  So, for example, if you choose a distance of 500m, and there are a couple of features that don't have any neighbors at 500m, but you also set a Number of Neighbors at 2...then for those features WITHOUT any neighbors at the specified distance band we'll increase the threshold to ensure those particular features have at least 2 neighbors.  You can only do it using the Generate Spatial Weights Matrix option, and then from Hot Spot Analysis you'll choose to Get Spatial Weights From File.  We often use this option when dealing with this issue.

Hope this helps!

Lauren Rosenshein
Geoprocessing Product Engineer

Check out the latest Spatial Statistics resources at http://esriurl.com/spatialstats

Hi Lauren,

Thanks so much for your suggestions.  It makes perfect sense.  I have a small question though.  For those no neighbor features, is choosing at least two neighbors as a parameter a rule of thumb?  I read somewhere else saying 8 neighbors as a rule of thumb.

Franky

[HuaqiYuan]

Hi Lauren,

Will it be in the next release of ArcGIS to include some routines to choose the spatial relationship and distance threshold for HSA automatically based on the evaluation of the raw data.

Franky