Help need for Local spatial statistics Moran's I and Geary's C -

Prev Go to previous topic

Go to next topic

Last Post 29 Oct 2018 12:08 PM by MariM

Help need for Local spatial statistics Moran's I and Geary's C

2 Replies

Sort:


You are not authorized to post a reply.

Author

Messages

anon

New Member

Posts:

21 Dec 2011 03:26 PM
I need come clarification regarding the Local Spatial Statistics tools(Getis-Ord G, Gearys C and Moran's I) From the literature ENVI cited(And from other sources) I found that "Moran’s I varies from -1 (negative autocorrelation) to 1 (positive autocorrelation). Geary’s c ranges from 0 (Positive Autpcorrelation) 1 (no auto correlation) and 2 (negative spatial autocorrelation)." However ENVI 4.5 is not obeying this rule, when I ran these methods on a raster file I have the following values Gearys's C " min:0.0155 and max:187.3784" Moran's I " min: -12.52 and max:201.62" Would you be able to provide some help on "Why this is happening" ? Thanks Naveen Anne

cao laura

New Member

Posts:1

28 Oct 2018 08:25 PM
Hello, I also met this question. Do you solve it?

MariM

Veteran Member

Posts:2396
Veteran Member

29 Oct 2018 12:08 PM
The global Moran's I value varies between -1 and 1, and the global Geary's C value typically varies between 0 and 2. However, the value constraints are not true for local Moran's I and local Geary's C. For a pixel cluster of similar values and significantly different from mean value, the pixel values of local Moran's I could be large. For pixels of strong edges or outliers, the pixel values of local Geary'C could be large. I also believe there was an update in a later version of ENVI (5.1 and later): https://www.harrisgeospatial.com/docs/LocalSpatialStatistics.html

You are not authorized to post a reply.