topic Re: getting Incremental Spatial Autocorrelation to work in 10.1 in Spatial Statistics Questions

getting Incremental Spatial Autocorrelation to work in 10.1

SimonWillis — Fri, 30 Nov 2012 21:21:18 GMT

hey all,

Using 10.1 sp1 & tutorial data I am able to successfully run Moran's I on my projected polygon shapefile data.
Some output:

Executing: SpatialAutocorrelation Calls_for_Service_Feb07_Gridshp Count_ GENERATE_REPORT ZONE_OF_INDIFFERENCE EUCLIDEAN_DISTANCE NONE 1000 #
Start Time: Fri Nov 30 15:11:16 2012
Running script SpatialAutocorrelation...

Global Moran's I Summary
Moran's Index:   0.056827
Expected Index: -0.002058
Variance:        0.000654
z-score:         2.303085
p-value:         0.021274

Distance measured in US_Feet

Writing html report....
so everything is fine here.

Instead of running this tool x times with different bands, I thought I'd try the incremental spatial autocorrelation but I can't get it to work in 10.1. I use the same dataset and the same beginning band I used in Moran's I. I get this error:

Executing: IncrementalSpatialAutocorrelation Calls_for_Service_Feb07_Gridshp Count_ 30 1000 100 EUCLIDEAN ROW_STANDARDIZATION # #
Start Time: Fri Nov 30 14:58:37 2012
Running script IncrementalSpatialAutocorrelation...

Calculating Number of Neighbors Per Increment....
ERROR 001388: None of the features have neighbors at the following distances: 1000.00, 1100.00, 1200.00, 1300.00, 1400.00, 1500.00, 1600.00, 1700.00, 1800.00, 1900.00, 2000.00, 2100.00, 2200.00, 2300.00, 2400.00, 2500.00, 2600.00, 2700.00, 2800.00, 2900.00, 3000.00, 3100.00, 3200.00, 3300.00, 3400.00, 3500.00, 3600.00, 3700.00, 3800.00, 3900.00
Completed script IncrementalSpatialAutocorrelation...
Failed to execute (IncrementalSpatialAutocorrelation).
Failed at Fri Nov 30 14:58:38 2012 (Elapsed Time: 1.00 seconds)

Using this dataset in 9.2 with the downloaded spatial stats toolbox works fine.

Can anyone shed light on why I can't get it to work in 10.1?
Thanks!

Re: getting Incremental Spatial Autocorrelation to work in 10.1

DavidRocha1 — Mon, 03 Dec 2012 17:38:20 GMT

I came here to make a similar thread.

What I've found is that the input parameter can detemine whether the tool fails or not. I am using meters, and 30 increments.

*All* failures I am getting are telling me there are no neighbors at those distances (which is wrong):
ERROR 001388: None of the features have neighbors at the following distances..

Leaving the initial value constant, if I set the increment to:
100 = fail
500 = fail
750 = works
1000 = works

I tried various values from 500-750 and even 725 failed.

So, I'd like to know why the script is failing depending on increment when I know my tracts have neighbors at at least one of the distances, the output is telling me that no features have any neighbors. Yet if I use increment of 750, I can see plenty of results with the range that the script errantly told me there are no neighbors.

Re: getting Incremental Spatial Autocorrelation to work in 10.1

SimonWillis — Mon, 03 Dec 2012 18:03:18 GMT

so have you ran moran's I at different increments and gotten correct results?
I am wondering if anyone can get incremental spatial auto-corr tool to run successfully & correctly in 10.1 - Can someone confirm please? (I can in 10.0 - sorry incorrectly mentioned 9.2 in my first post).
Could this be a bug?