+ Reply to Thread
Results 1 to 6 of 6

Thread: better clustering

  1. #1
    Points: 2,455, Level: 30
    Level completed: 4%, Points required for next Level: 145

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    better clustering




    Hi,
    I have some 40 to 50 arrays of numerical values i want to divide into some set groups by their trend can u guys suggest me a better approach.

    Thanks in advance,
    Raj

  2. #2
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts
    You might need to provide a bit more detail about what you're trying to do here! What are you analysing? What kind of groups do you want to divide the data into?

  3. #3
    Points: 2,455, Level: 30
    Level completed: 4%, Points required for next Level: 145

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Please find the attached file.
    the first column values are some codes and the remaining values in the next columns are the values of those codes now i want to group them into some groups based on the values present after the second column onwards.

    Thanks In advance,
    Raj.
    Attached Files

  4. #4
    Points: 2,422, Level: 29
    Level completed: 82%, Points required for next Level: 28

    Posts
    13
    Thanks
    0
    Thanked 0 Times in 0 Posts
    You will need to explain what these data are for us to help. What are the columns? What are the codes? The data set is small and will be easy to work with, we just need to know what is what!

  5. #5
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts
    Thanks for providing the data... In addition to teh physiologyfellow's requests, I'm also not really clear on what basis you're wanting to group the variables on? To use statistical techniques you kinda need a clear idea of what you're wanting to do

  6. #6
    Points: 2,455, Level: 30
    Level completed: 4%, Points required for next Level: 145

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Thanks for ur replies
    In the attached Excel i want to group the codes based on the values of the next 5 columns and one more thing this results are going to generate dynamically so there won't be any limit for codes and there won't be any limit for the values of those codes in this case

    if u look at row 24 which is having 4 values based on which we need to decide trend for this code in this case the number of values are 4 some times it may be 5 or 10 or any thing.

    N0110 6 5.727272727 4.636363636 3.454545455

    here No110 code trend is continuously decreasing like that there might be some other codes which will follow this trend which i want group as one cluster like that there might be so many clusters with unique behavior i want to make that many clusters.

    Thanks in advance,
    Raj.

+ Reply to Thread

           




Similar Threads

  1. Robust Clustering?
    By afclark82 in forum Stata
    Replies: 0
    Last Post: 04-07-2011, 12:09 PM
  2. Clustering time series
    By abanero in forum R
    Replies: 0
    Last Post: 08-26-2010, 02:38 AM
  3. Clustering in SPSS
    By weakatstats in forum Probability
    Replies: 4
    Last Post: 08-18-2010, 04:36 PM
  4. Clustering?
    By Kam36 in forum Psychology Statistics
    Replies: 0
    Last Post: 06-03-2008, 05:06 PM
  5. clustering test
    By german in forum Statistics
    Replies: 1
    Last Post: 11-29-2006, 09:00 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats