Global Moran's I / Moran scatter plot with islands

lixun910 commented 6 years ago

since the isolated value comes into play in computing the mean and variance. GeoDa only supports row-standardized weights in the global Moran. Our solution is to eliminate the isolates from both numerator and denominator in the computation. The more fundamental point is that isolates do not contribute to any “spatial interaction” so they should not be included in any measure of spatial autocorrelation. For the local Moran etc. in GeoDa, this is already the case.

issue warning “Moran scatter plot not supported when isolates are present”
if user proceeds, we trim the island locations from the analysis, make sure the standardized values do not include the islands in the computation of mean and variance, and generate the scatter plot (with a warning that isolates have been removed)

YidanJ-Wu commented 6 years ago

Would this affect clustering algorithms that use weights files (SKATER, REDCAP)?

lixun910 commented 6 years ago

No, this won’t affect clustering algorithms. I think we could apply different strategies, e.g. (ask user to choose) either connect islands to nearest to make a complete graph or remove islands in the background, to handle islands in clustering methods.

On May 27, 2018, at 9:40 AM, YidanJ-Wu notifications@github.com wrote:

Would this affect clustering algorithms that use weights files (SKATER, REDCAP)?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/GeoDaCenter/geoda/issues/1654#issuecomment-392346099, or mute the thread https://github.com/notifications/unsubscribe-auth/ACT8T9u8SPDX4j_i0K689Pm4QVKp51iqks5t2tcZgaJpZM4UNQz9.

Temisesba commented 3 years ago

como puedo hacer lo mismo en rgeoda al hacer el scatter plot?

jkoschinsky commented 3 years ago

Hi, this is not in rgeoda because it is already implemented in R's spdep. See here, e.g.: https://mgimond.github.io/Spatial/spatial-autocorrelation-in-r.html I Spatial autocorrelation in R | Intro to GIS and Spatial Analysis - GitHub Pageshttps://mgimond.github.io/Spatial/spatial-autocorrelation-in-r.html The data object consists of a SpatialPolygonsDataFrame vector layer, s1, representing income and education data aggregated at the county level for the state of Maine.. The spdep (Roger S. Bivand 2013) package used in this exercise makes use of sp objects including SpatialPoints and SpatialPolygons classes. For more information on converting to/from this format revert back to the Reading and ... mgimond.github.io

Julia Koschinsky, Ph.D. Executive Director and Senior Research Associate Center for Spatial Data Science University of Chicago https://spatial.uchicago.edu/

https://spatial-data-science-conference.com/

From: Temisesba @.> Sent: Tuesday, October 5, 2021 6:21 PM To: GeoDaCenter/geoda @.> Cc: Julia Koschinsky @.>; State change @.> Subject: Re: [GeoDaCenter/geoda] Global Moran's I / Moran scatter plot with islands (#1654)

como puedo hacer lo mismo en rgeoda al hacer el scatter plot?

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHubhttps://github.com/GeoDaCenter/geoda/issues/1654#issuecomment-935047052, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ADILY7ZSOCBMZPRBIBJNJA3UFOB6VANCNFSM4FBVBT6Q. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

GeoDaCenter / geoda

Global Moran's I / Moran scatter plot with islands #1654