+ Reply to Thread
Results 1 to 11 of 11

Thread: Nonparametric methods

  1. #1
    Points: 3,621, Level: 37
    Level completed: 81%, Points required for next Level: 29
    askazy's Avatar
    Location
    Holand
    Posts
    160
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Nonparametric methods




    If I have a sample with the size of the right and left arms of each of these people.So I have two dependent groups, since these are the same person, right? I I can apply a Wilcoxon test to see if the two groups are equal, or if there is a difference between these?
    The difference between stupidity and genius is that genius has its limits.
    "Albert Einstein"

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Nonparametric methods

    askazy, you driving me crazy with the Wilcoxon. Try to add which test in particular, since his name is associated with many tests.

    Yes, you could subtract the two paired values and use the Wilcoxon sign rank test (on the differences). What is your sample size and data distribution. Many programs will conduct an exact test when the sample sizes are small (e.g., in SAS, n-value < 20), also some people feel the central limit theorem comes into play as the sample size increases.

    This is a better dataset idea than the previous proposed sets!
    Stop cowardice, ban guns!

  3. #3
    Points: 3,621, Level: 37
    Level completed: 81%, Points required for next Level: 29
    askazy's Avatar
    Location
    Holand
    Posts
    160
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Re: Nonparametric methods

    Quote Originally Posted by hlsmith View Post
    askazy, you driving me crazy with the Wilcoxon. Try to add which test in particular, since his name is associated with many tests.

    Yes, you could subtract the two paired values and use the Wilcoxon sign rank test (on the differences). What is your sample size and data distribution. Many programs will conduct an exact test when the sample sizes are small (e.g., in SAS, n-value < 20), also some people feel the central limit theorem comes into play as the sample size increases.

    This is a better dataset idea than the previous proposed sets!

    So the question is do I have to prepare a study where I use this test. Data can not have normal distribution. How have to collect the data I think of using a sample of size 15 or 20.
    The ideas that I had to apply this test following these standards are:

    These are the ideas that I had, but I'm sure if they fit in this test. Because exercise says to compare two groups, and do not know if these ideas fit together like two groups.
    Last edited by askazy; 12-12-2014 at 03:18 PM.
    The difference between stupidity and genius is that genius has its limits.
    "Albert Einstein"

  4. #4
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Nonparametric methods

    So you will have a dataset with difference values. Now you want to stratify the data set and test additional hypotheses. I am guessing using Wilcoxon rank sum.

    Not sure if your Wilcoxon sign rank test is not significant if that will indirectly lower the potential power for subsequent tests, unless all of the -'s and +'s signs get group together in your subsequent stratifications.
    Stop cowardice, ban guns!

  5. #5
    Points: 3,621, Level: 37
    Level completed: 81%, Points required for next Level: 29
    askazy's Avatar
    Location
    Holand
    Posts
    160
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Re: Nonparametric methods

    Quote Originally Posted by hlsmith View Post
    So you will have a dataset with difference values. Now you want to stratify the data set and test additional hypotheses. I am guessing using Wilcoxon rank sum.

    Not sure if your Wilcoxon sign rank test is not significant if that will indirectly lower the potential power for subsequent tests, unless all of the -'s and +'s signs get group together in your subsequent stratifications.
    Look realizing that third option that I presented obtained the following result
    Last edited by askazy; 12-12-2014 at 03:04 PM.
    The difference between stupidity and genius is that genius has its limits.
    "Albert Einstein"

  6. #6
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts
    Stop cowardice, ban guns!

  7. The Following User Says Thank You to hlsmith For This Useful Post:

    askazy (12-12-2014)

  8. #7
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Nonparametric methods

    A minor point. SAS will run an exact test with any sample size. It just takes longer. I think you could run an exact test in a reasonable time with a lot more than 20 cases.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  9. #8
    Points: 3,621, Level: 37
    Level completed: 81%, Points required for next Level: 29
    askazy's Avatar
    Location
    Holand
    Posts
    160
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Re: Nonparametric methods

    I have doubts about the state the hypotheses of my test. If my test is about the size difference between the arms of a person, then the chances should be

    H_0: There is no difference between the sizes of arms
    H_1: There is a difference between the sizes of arms

    I am confused by this because in some references I read, the assumptions are set based on the median.

    What is the test statistic should I use for a population of size 20?
    I think not give to approach the normal, then I use the Wilcoxon table to determine the critical region of the right sided test?
    The difference between stupidity and genius is that genius has its limits.
    "Albert Einstein"

  10. #9
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Nonparametric methods

    I typically run the non-parametrics based on median since extreme outliers can provide an off measure of central tendancy.
    Stop cowardice, ban guns!

  11. The Following User Says Thank You to hlsmith For This Useful Post:

    askazy (12-16-2014)

  12. #10
    Points: 3,621, Level: 37
    Level completed: 81%, Points required for next Level: 29
    askazy's Avatar
    Location
    Holand
    Posts
    160
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Re: Nonparametric methods

    Quote Originally Posted by hlsmith View Post
    I typically run the non-parametrics based on median since extreme outliers can provide an off measure of central tendancy.
    What is the correct way to state the hypotheses:
    H_0:m_0=m
    H_1:m_0=m
    Where m is the median, or
    H_0:There is no difference between the size of the arms
    H_1:There is difference between the size of the arms.

    Only that was missing to complete the job.
    Last edited by askazy; 12-13-2014 at 12:12 PM.
    The difference between stupidity and genius is that genius has its limits.
    "Albert Einstein"

  13. #11
    Points: 3,621, Level: 37
    Level completed: 81%, Points required for next Level: 29
    askazy's Avatar
    Location
    Holand
    Posts
    160
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Re: Nonparametric methods


    Solved, thank you.
    The difference between stupidity and genius is that genius has its limits.
    "Albert Einstein"

+ 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