Loading ...
Sorry, an error occurred while loading the content.

AI-GEOSTATS: Cross-correlation 2D

Expand Messages
  • Joseph Le Cuziat
    Hi all. I m looking for the best way to put into evidence spatial structure correlation between two maps. I ve computed local density maps from two different
    Message 1 of 3 , Apr 2, 2004
    • 0 Attachment
      Hi all.

      I'm looking for the best way to put into evidence spatial structure
      correlation between two maps.

      I've computed local density maps from two different point pattern (herd
      and bird distribution over a study area) using kernel smoother, and I
      want now to compare these local density to see if there is some evidence
      for correlation in their spatial structure.

      Cross-correlation seems to do such a job, and I found it's often use to
      compare time series.
      But I was not able to find software or package that can do the job in 2D
      (I usually use R), since I'm working in space and not time...

      1/ Do you think Cross-correlation is a good way to compare such density
      maps ?

      2/ Does anyone knows what software (or R package) can do the job ?

      Many thanks in advance.

      Sincerely.
      Joseph.


      --
      Joseph Le Cuziat <jlecuziat@...>
      IMEP - Institut Mediterraneen d'ecologie et de Paleoecologie


      --
      * To post a message to the list, send it to ai-geostats@...
      * As a general service to the users, please remember to post a summary of any useful responses to your questions.
      * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
      * Support to the list is provided at http://www.ai-geostats.org
    • Roger Bivand
      ... Well, if you still have the point patterns, you might find that you can attack your problem directly using a bivariate K-function, for example k12hat() in
      Message 2 of 3 , Apr 2, 2004
      • 0 Attachment
        On Fri, 2 Apr 2004, Joseph Le Cuziat wrote:

        > Hi all.
        >
        > I'm looking for the best way to put into evidence spatial structure
        > correlation between two maps.
        >
        > I've computed local density maps from two different point pattern (herd
        > and bird distribution over a study area) using kernel smoother, and I
        > want now to compare these local density to see if there is some evidence
        > for correlation in their spatial structure.
        >
        > Cross-correlation seems to do such a job, and I found it's often use to
        > compare time series.
        > But I was not able to find software or package that can do the job in 2D
        > (I usually use R), since I'm working in space and not time...
        >
        > 1/ Do you think Cross-correlation is a good way to compare such density
        > maps ?
        >
        > 2/ Does anyone knows what software (or R package) can do the job ?

        Well, if you still have the point patterns, you might find that you can
        attack your problem directly using a bivariate K-function, for example
        k12hat() in the R splancs package; I think Kcross() in the R spatstat
        package will do this too. I think the density maps will be inducing local
        dependence in the grid cells, so trying to compare them might be OK just
        as a scatterplot, but the correlation would be driven by the kernel and
        bandwidth (large bandwidth giving more autocorrelation, I guess). The
        k12hat() approach is only using the initial data, so it doesn't depend on
        your choice of kernel or bandwidth.

        >
        > Many thanks in advance.
        >
        > Sincerely.
        > Joseph.
        >
        >
        >

        --
        Roger Bivand
        Economic Geography Section, Department of Economics, Norwegian School of
        Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
        Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
        e-mail: Roger.Bivand@...



        --
        * To post a message to the list, send it to ai-geostats@...
        * As a general service to the users, please remember to post a summary of any useful responses to your questions.
        * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
        * Support to the list is provided at http://www.ai-geostats.org
      • mrufino@cmima.csic.es
        Hello Joseph, We used spatial Mantel test for finding the correlation between two maps. You can do it all with R (we also use R)- you have three correlation
        Message 3 of 3 , Apr 4, 2004
        • 0 Attachment
          Hello Joseph,

          We used spatial Mantel test for finding the correlation between two maps.
          You can do it all with R (we also use R)- you have three correlation
          coefficients available (spearman, pearson and kendal).
          You can also redraw the effect of a third variable, by partial mantel test.
          If you want more details about how we did it, write me, please.
          I recently found a work from Pebesma with brid densities, which do another
          approach.... really nice. But I have not tried it yet.

          Hope this helps,
          cheers
          Marta


          Quoting Joseph Le Cuziat <jlecuziat@...>:

          > Hi all.
          >
          > I'm looking for the best way to put into evidence spatial structure
          > correlation between two maps.
          >
          > I've computed local density maps from two different point pattern (herd
          > and bird distribution over a study area) using kernel smoother, and I
          > want now to compare these local density to see if there is some evidence
          > for correlation in their spatial structure.
          >
          > Cross-correlation seems to do such a job, and I found it's often use to
          > compare time series.
          > But I was not able to find software or package that can do the job in 2D
          > (I usually use R), since I'm working in space and not time...
          >
          > 1/ Do you think Cross-correlation is a good way to compare such density
          > maps ?
          >
          > 2/ Does anyone knows what software (or R package) can do the job ?
          >
          > Many thanks in advance.
          >
          > Sincerely.
          > Joseph.
          >
          >
          > --
          > Joseph Le Cuziat <jlecuziat@...>
          > IMEP - Institut Mediterraneen d'ecologie et de Paleoecologie
          >
          >
          > --
          > * To post a message to the list, send it to ai-geostats@...
          > * As a general service to the users, please remember to post a summary of any
          > useful responses to your questions.
          > * To unsubscribe, send an email to majordomo@... with no subject and
          > "unsubscribe ai-geostats" followed by "end" on the next line in the message
          > body. DO NOT SEND Subscribe/Unsubscribe requests to the list
          > * Support to the list is provided at http://www.ai-geostats.org
          >




          -------------------------------------------------
          This mail sent through IMP: http://horde.org/imp/

          --
          * To post a message to the list, send it to ai-geostats@...
          * As a general service to the users, please remember to post a summary of any useful responses to your questions.
          * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
          * Support to the list is provided at http://www.ai-geostats.org
        Your message has been successfully submitted and would be delivered to recipients shortly.