Notices
Results 1 to 6 of 6

Thread: Confidence Interval

  1. #1 Confidence Interval 
    Forum Freshman
    Join Date
    Jan 2012
    Posts
    16
    Hi,

    Sample size iz n=9. I have a values in percentage which indicate a certain deviation (either positive or negative). For example:

    -1.25%, +1.63% ... etc.

    I get the 95% CI that has lower bound negative and upper bound positive. Can this be true or something is wrong?

    Many thanks.


    Last edited by ivanf; November 9th, 2013 at 05:02 PM.
    Reply With Quote  
     

  2.  
     

  3. #2  
    Forum Radioactive Isotope MagiMaster's Avatar
    Join Date
    Jul 2006
    Posts
    3,440
    Without knowing what those percentages represent, I couldn't really say, but it's certainly possible for some ways of calculating confidence intervals to put the range out of bounds. You could try recalculating things in a logarithmic domain if you know your numbers must be positive.


    Reply With Quote  
     

  4. #3  
    Forum Freshman
    Join Date
    Jan 2012
    Posts
    16
    Percentages represent the change in price. The values are: -1.25%, +1.63%, +0,22%, -0,56%, +21,54%, -1,83%, -1,63%, -0,7%, -0,37%.

    My results are 95% CI = 1.89% +/- 5.71%. Is this correct?

    P.S. I have used t distribution.

    Best regards,
    Reply With Quote  
     

  5. #4  
    Forum Radioactive Isotope MagiMaster's Avatar
    Join Date
    Jul 2006
    Posts
    3,440
    Looks about right to me, or at least I got the same answer (though 5.719 should probably be rounded to 5.72). Since prices can certainly drop (and you even have negative samples), a negative lower bound is certainly reasonable.
    Reply With Quote  
     

  6. #5  
    Forum Freshman
    Join Date
    Jan 2012
    Posts
    16
    Thank you for your effort.

    One more thing, how to interpret the results given the small sample size?

    Best regards.
    Reply With Quote  
     

  7. #6  
    Forum Radioactive Isotope MagiMaster's Avatar
    Join Date
    Jul 2006
    Posts
    3,440
    That would depend on how you got the sample I think. Generally though, you would expect another sample to fall within those bounds 95% of the time.
    Reply With Quote  
     

Similar Threads

  1. Confidence Interval; Statistics Homework Help
    By marcusclayman in forum Mathematics
    Replies: 3
    Last Post: December 10th, 2010, 09:36 PM
  2. Confidence and joy
    By Charlatan in forum Behavior and Psychology
    Replies: 20
    Last Post: November 12th, 2010, 11:35 AM
  3. Matching tolerance interval
    By Faron in forum Mathematics
    Replies: 7
    Last Post: May 29th, 2008, 08:28 AM
  4. Confidence in reason
    By coberst in forum Philosophy
    Replies: 7
    Last Post: October 27th, 2007, 02:24 PM
  5. Galactic Geologic Interval Hypothesis
    By morbas in forum Pseudoscience
    Replies: 12
    Last Post: March 16th, 2007, 05:56 AM
Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •