+ Reply to Thread
Results 1 to 6 of 6

Thread: Outliers in a normal Distribution

  1. #1
    Points: 2,983, Level: 33
    Level completed: 56%, Points required for next Level: 67

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Outliers in a normal Distribution




    So far, all I've been able to figure out is the IQR in percentage is 50%. This is the problem.

    The percent of the observations that are classified as outliers by the 1.5x IQR rule is the same in any normal distribution. What is this percent? Show your method clearly.

    I really don't have a clue how to begin to solve this. If someone can get me started, I would much appreciate it.

  2. #2
    Points: 2,804, Level: 32
    Level completed: 36%, Points required for next Level: 96

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts
    I too am doing this exact same question right now, and I have no idea how to do it. If anyone can help out, it would be greatly appreciated.

  3. #3
    Points: 2,988, Level: 33
    Level completed: 59%, Points required for next Level: 62

    Posts
    4
    Thanks
    0
    Thanked 1 Time in 1 Post
    I'm sorry fellas but there doesn't seem to be anyone answering in this forum, only people doing questions. I'm new to statistics forums, so maybe this is the case in a such specilized context as statistics...

  4. #4
    Points: 2,848, Level: 32
    Level completed: 66%, Points required for next Level: 52

    Posts
    28
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Think of a normal density plot. What is the range of x values which would contain 50% of the values centered on the mean? Now, how many values would be contained in +/- 1.5 IQR and what is the x range? How many values are left over?

    Hints: think z-scores; boxplots; the meaning of 1.5 IQR

    Sorry if this seems cryptic but, really, it doesn't get much simpler than this.
    Last edited by DAV; 10-04-2008 at 03:29 PM. Reason: Slight clarification

  5. #5
    Points: 2,794, Level: 32
    Level completed: 30%, Points required for next Level: 106

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts
    The interquartile range is

    Q_3 - Q_1

    Try to write both quartiles in terms of the mean and standard deviation of the distribution, and that will give you the spread about the mean in language the normal distribution is designed to use

  6. #6
    Points: 2,861, Level: 32
    Level completed: 74%, Points required for next Level: 39

    Location
    Pennsylvania
    Posts
    14
    Thanks
    0
    Thanked 0 Times in 0 Posts

    outliers in normal distribution


    look up Q1 and Q3 in the Stand.Norm.Table they are +/- 0.675. Then take Q3 -Q1 and you get 1.35. Then use Tukey's outlier rule and multiply by 1.5 to get 2.025. Add to the Q3 and get cutoffs of z = 2.70. Then look up the area to the right 0.0034 and double, multiply by 100 and get 0.69%

+ Reply to Thread

           




Similar Threads

  1. Replies: 2
    Last Post: 05-11-2011, 01:35 PM
  2. Normal distribution, outliers, and the median
    By musterseed in forum Statistics
    Replies: 5
    Last Post: 06-11-2010, 03:19 AM
  3. Is a subset of a normal distribution normal?
    By bordonbert in forum Probability
    Replies: 4
    Last Post: 10-29-2009, 03:54 AM
  4. Normal? Bimodal normal? Some other distribution?
    By statsCZ in forum Statistical Research
    Replies: 4
    Last Post: 01-28-2009, 07:31 AM
  5. Replies: 0
    Last Post: 01-23-2008, 10:32 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