+ Reply to Thread
Results 1 to 4 of 4

Thread: General method for data analysis

  1. #1
    Points: 55, Level: 1
    Level completed: 10%, Points required for next Level: 45

    Posts
    3
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Post General method for data analysis




    Dear community,

    maybe you can help me. I have a lots of data like this format:

    Person/points/work/power/sex
    Person1/75/10/120/m
    Person2/39/5/100/w
    Person3/46/7/150/m
    ...

    Well, these a special values and I try to find for example some "unusually" persons, maybe in a graphic. Maybe it sounds a bit weird, but I do not know what persons (or person groups) I try to find, so I canīt set a special goal or structure.
    So I need a more general method to give me a structure and a way to find some unusally or common things.
    I tried a method called Principal component analysis (PCA) and it seems this gives a nice pattern to identification special person-groups, but I need to find a better method for this.
    Does anyone knows more general methods I can use (or give me a hint what to do here), if I still donīt know "what I try to find"?

    Regards
    Dotana

  2. #2
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 639 Times in 601 Posts

    Re: General method for data analysis

    I try to find for example some "unusually" persons (...)
    identification special person-groups
    So, what do you want to identify, single unusual persons (multivarate outliers
    http://tinyurl.com/pozwkb2), or one or more clusters not-so-extremely-unusual
    persons?

    With kind regards

    K.

  3. The Following User Says Thank You to Karabiner For This Useful Post:

    Dotana (11-13-2013)

  4. #3
    Points: 55, Level: 1
    Level completed: 10%, Points required for next Level: 45

    Posts
    3
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: General method for data analysis

    Quote Originally Posted by Karabiner View Post
    So, what do you want to identify, single unusual persons (multivarate outliers
    http://tinyurl.com/pozwkb2), or one or more clusters not-so-extremely-unusual
    persons?

    With kind regards

    K.
    Hello Karabiner,

    both seems to be interesting for me.
    I ll follow your link and hope to find something, thanks for that.
    You mentioned 'clusters not-so-extremely-unusual ', do you have a method/technique in mind?

    Regards
    Dotana

  5. #4
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 639 Times in 601 Posts

    Re: General method for data analysis


    There are several cluster analysis algorithms, although the mixture
    of interval scaled and categorical variables could represent a problem
    for such methods.

    With kind regards

    K.

+ Reply to Thread

           




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