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

GEOSTATS: test for uniform planting site?

Expand Messages
  • Ali_Alaa@seo.state.nm.us
    You can use a KNN (Kth Nearest Neighbor) bootstrap. Perform the following (considering your example): 1) let knn = sqrt(30) = 5 or 6 (you may define any other
    Message 1 of 2 , Mar 26, 1997
      You can use a KNN (Kth Nearest Neighbor) bootstrap. Perform the
      following (considering your example):
      1) let knn = sqrt(30) = 5 or 6 (you may define any other value)
      2) Visit the 30 locations along a random path
      3) At location i, consider the nearest 6 points. Calculate a re-sampling
      probability of each of these 6 points such that the closer the point to
      point i, the higher the probability. (use the inverse distance and
      normalize, or use Kernel function)
      4) Re-sample point i according the probabilities caclulated above
      5) proceed along the random path to re-sample all points and obtain 1
      realization.
      6) repeat many times with different random paths (10000....).
      7) Calculate the fraction of F, and S (if you are working with growth
      height, you can categorize it) from the obtained realizations.

      Good luck..
      A. Ali
      _________________________________________________________
      A. Ali, Ph.D., P.E.
      New Mexico State Engineer Office
      P.O.Box 25102, Santa Fe, NM 87504
      e-mails: aali@...; or aali@...
      Web: http://www.engineering.usu.edu/Departments/cee/Faculty/ulall/
      Phone: 505-827-6125 Fax: 505-827-6188
      _________________________________________________________


      > I am looking for a test that planting location
      > influences seedling growth. For example,
      > 30 seedlings are planted in a uniform medium.
      > The S represent "slow growers" and the F
      > represent "fast growers".
      >
      > F S S F S F
      > F F S F S S
      > S S F S F S
      > S S S F S F
      > F S F S F S
      >
      > The above is a simplified example as
      > I have height growth for each seedling
      > and about 600 seedlings.
      >
      > One approach would be a jacknife or bootstrap
      > type approach of randomly assigning each seedling
      > growth to each location and repeating the randomiztion
      > trials 1,000 times. Then compare the original distribution
      > with the 1,000 randomization trial distributions. But
      > how?
      >
      > The null hypothesis is that the site is uniform and
      > all variablity of seedling growth is due to factors
      > other than seedling location.
      >
      > Any suggestions would be appreciated. I will post
      > a summary on April 15, 1997.
      >
      > Thank you.
      >
      > Dave Verbyla
      > Associate Professor of GIS/Remote Sensing
      > --
      > *To post a message to the list, send it to ai-geostats@....
      > *As a general service to list users, please remember to post a summary
      > of any useful responses to your questions.
      > *To unsubscribe, send email to majordomo@... with no subject
      and
      > "unsubscribe ai-geostats" in the message body.
      > DO NOT SEND Subscribe/Unsubscribe requests to the list!
      >

      --
      *To post a message to the list, send it to ai-geostats@....
      *As a general service to list users, please remember to post a summary
      of any useful responses to your questions.
      *To unsubscribe, send email to majordomo@... with no subject and
      "unsubscribe ai-geostats" in the message body.
      DO NOT SEND Subscribe/Unsubscribe requests to the list!
    Your message has been successfully submitted and would be delivered to recipients shortly.