+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 16 to 26 of 26

Thread: tests for normal data

  1. #16
    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: tests for normal data




    If you accept that the power is weak, then there certainly is no rule against using them. If there was they would not exist Anderson-Darling is also used a lot.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  2. #17
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Posts
    135
    Thanks
    18
    Thanked 2 Times in 2 Posts

    Re: tests for normal data

    Quote Originally Posted by noetsi View Post
    If you accept that the power is weak, then there certainly is no rule against using them. If there was they would not exist Anderson-Darling is also used a lot.
    I know about the power, but I don't have more data. I'm trying to do the best that is possible with the data that I have...

    Thank you very much

  3. #18
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Posts
    135
    Thanks
    18
    Thanked 2 Times in 2 Posts

    Re: tests for normal data

    Hi again.
    Anf if my data is
    Code: 
     0   0   0   0   0   1   1   1   1   1  1   1   1   0   0   0  0  1  0  0  0  0  0   1   1   1   1   1   1  1   1   1   1  1   1   1   1   1   1   1   1  1  1  1  1  1  1  1
    it's correct to apply a shapiro.test to test the normality of data?

  4. #19
    Devorador de queso
    Points: 95,889, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,937
    Thanks
    307
    Thanked 2,630 Times in 2,246 Posts

    Re: tests for normal data

    Here's a hint - if the only values your data can take are 0 and 1 then the data itself is NOT normally distributed...
    I don't have emotions and sometimes that makes me very sad.

  5. #20
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Posts
    135
    Thanks
    18
    Thanked 2 Times in 2 Posts

    Re: tests for normal data

    Hi.

    I can have values like

    Code: 
     0   0   0   0   0   1   2   4   1   1  4   2  1   0   0   0  0  1  44  0  25  0  0   1   1   1   1   1   1  1   1   1   1  1   1   1   1   1   1   1   1  1  1  1  1  1  1  1
    It's good to apply a shapiro.test?

  6. #21
    Devorador de queso
    Points: 95,889, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,937
    Thanks
    307
    Thanked 2,630 Times in 2,246 Posts

    Re: tests for normal data

    It doesn't really matter. Your data isn't normal if it's like that. Once again I'll mention that it looks like you're working with count data. Maybe you should just look into a method that's appropriate for for you data instead of trying to figure out if your data is "normal" (Because it's not)
    I don't have emotions and sometimes that makes me very sad.

  7. The Following 2 Users Say Thank You to Dason For This Useful Post:

    GretaGarbo (05-06-2013), hlsmith (05-06-2013)

  8. #22
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Posts
    135
    Thanks
    18
    Thanked 2 Times in 2 Posts

    Re: tests for normal data

    Quote Originally Posted by Dason View Post
    It doesn't really matter. Your data isn't normal if it's like that. Once again I'll mention that it looks like you're working with count data. Maybe you should just look into a method that's appropriate for for you data instead of trying to figure out if your data is "normal" (Because it's not)

    But the problem is that I have a lot of data like this. And I know to test the normality for each one, because after I will apply the test automatically. I can't look for 20000 groups of data to apply the test... You understand?

  9. #23
    Devorador de queso
    Points: 95,889, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,937
    Thanks
    307
    Thanked 2,630 Times in 2,246 Posts

    Re: tests for normal data

    If this is count data then for a large enough mean the data will probably be approximately fairly well by a normal distribution. But like I mentioned before if this is count data then why are you so opposed to using a method that is more appropriate for count data? Something like a negative binomial regression or an overdispersed poisson regression. You might even have some zero inflation that you want to deal with.

    But really knowing more about your data is the first place to start.
    I don't have emotions and sometimes that makes me very sad.

  10. #24
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Posts
    135
    Thanks
    18
    Thanked 2 Times in 2 Posts

    Re: tests for normal data

    I'm not opposed to using other methods, I only didn't understand how can I use that. If you could help me how to do, thank you very much.
    My pproblem is that I need to do a program to do that becausa I hava a lot of rows with data (a lot of groups of data) and I need to create some rules to apply my tests.
    Thank you very much.

  11. #25
    Devorador de queso
    Points: 95,889, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,937
    Thanks
    307
    Thanked 2,630 Times in 2,246 Posts

    Re: tests for normal data

    Are you working at a university? Could you possibly contact a statistician in your area and try to arrange a face to face? You've asked the same question many times and our advice has been the same every time. We're more than happy to help if you put some effort in but you just seem to ask the same question over and over. I offered up a few alternatives but we aren't going to run the analysis for you. If you have a local statistician you could meet with then that would be best since it would be a lot easier for them to get an idea about your data and help you out.
    I don't have emotions and sometimes that makes me very sad.

  12. #26
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Posts
    135
    Thanks
    18
    Thanked 2 Times in 2 Posts

    Re: tests for normal data


    Ok, thank you

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2

           




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