Question with the Tool: Cluster and Outlier Analysis (Anselin Local Moran's I) (Spatial Statistics)

[CanZhang]

Hello everyone!

I have a statistic problem with the tool "Cluster and Outlier Analysis (Anselin Local Moran's I) (Spatial Statistics)".

On the web site ArcGIS Help 10.1  is the formula of Local Moran 'I.

There is a parameter Si2. Based on the literature, this parameter muss be a variance. But the formula of this parameter Si2 is not the same as a normal variance.

It's hard to describe the problem only in text, so I've made a PDF-Daten in attachment for better understanding.

Is there any special mathematical sinn of this formula of Si2 in ArcGIS or is that just wrong??

Is the calculation in the toolbox based on this wrong formula also injustice?

I'm looking forward to your answer! Thanks a lot!

[CanZhang]

anyone can help?

[GISSupport3]

G'day

You have posted to the wrong group / community / place

[Kylie]

I've moved your post to the Spatial Statistics space instead of the space for the Collector for ArcGIS app. That will help the right people see and answer your question.

[CanZhang]

Thank you!

[MarkJanikas · Accepted solution]

Thanks for spotting this Can... it was in fact, an error in our documentation.  Rest assured that the values are calculated correctly in the code... but you can always check yourself as it is in Python :-).  We will update the documentation ASAP.  Again, many thanks.

Mark Janikas

Product Developer

[CanZhang]

thank you for the answer!

Can you please tell me, where can I see the Python-code?

[MarkJanikas]

The $ARCHOME\ArcToolbox\Scripts directory contains all of the code for the Python Tools.  "LocalI.py" is the one you are looking for.  Best wishes,

MJ

[MarkJanikas]

Correction:  "LocalMoran.py"

[CanZhang]

thank you so much!