I need to calculate the Global and Local Moran indices in R using a variable distance as threshold: I have a SpatialPointsDataFrame with almost 300 points and I want to calculate the Global Moran index using 4 different distances (e.g 5 km - 10 km - 50 km - 100 km).
I read this discussion https://www.researchgate.net/post/Which_package_in_R_could_be_used_to_perform_Morans_I_test_for_spatial_autocorrelation
and I know the 'ape' and 'spdep' packages but it seems that no adjustment can be done concerning the spatial width to be considered...
Thanks
Try to use the "lisa" function and "speciesLisa" function from "usdm" (Uncertainty Analysis for Species Distribution Models) package by Babak Naimi.
https://cran.r-project.org/web/packages/usdm/index.html
Perhaps mantel.correlog function, which is part of the R vegan package can help.
http://cc.oulu.fi/~jarioksa/softhelp/vegan/html/mantel.correlog.html
Chris
Hi Maurizio,
the 'spatial width' or maximum distance of points to be considered is typically specified in the spatial weight matrix for Moran's I. Depending on your application you may want to use binary weights or something like squared inverse euclidean distance. If you want to calculate Moran's I for a maximum distance of e.g. 5km you simply need to assign a weight of 0 to all 'edges' between any two individual points further apart than 5km. Using spdep you will supply the spatial weights matrix as a listw object, so you need to specify the correct weights when you construct that listw object. You may also find useful functions in package 'adespatial'. Hope that helps,
Sebastian
Dear Maurizio,
If you don't mind to switch to Python, the PySAL library provides ready-to-use tools to compute both Global and Local Moran's I starting from a dataset of points.
I'm currently working on an experimental Python plugin for QGIS which includes an easy optimization to test spatial autocorrelation at different threshold distances. Code is available on GitHUB at the link below.
I remain available if you need further details or you are simply curious about.
Best,
Daniele
https://github.com/danioxoli/HotSpotAnalysis_Plugin
For anyone working on this and similar problems, I have found the spdep R package very helpful.
https://cran.r-project.org/web/packages/spdep/spdep.pdf
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology