Welcome to the NavList Message Boards.

NavList:

A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding

Compose Your Message

Message:αβγ
Message:abc
Add Images & Files
    or...
       
    Reply
    Re: Rejecting outliers: was: Kurtosis.
    From: George Huxtable
    Date: 2011 Jan 2, 17:27 -0000

    Marcel Tschudin wrote-
    |
    | For any symmetrical distribution mean, median and mode are identical.
    | If you have good reason to believe that the measured data are expected
    | to be symmetrically distributed, the median can be used. A
    | considerable difference between mean and median indicates either the
    | existance of outliers or a the possible existance of a skew
    | distribution. In "normal" datasets there is no a great difference
    | between the mean and the median. In small datasets an outlier
    | contributes too much. One outlier within e.g. 6 data contributes 17%
    | (if not weighted) whereas the outlier may in reality have a much lower
    | probability. The median thus helps to "correct" the influence of
    | outliers in small datasets.
    
    It is true that in a symmetrical distribution, the mean and the median are
    the same, at the centre of symmetry. But we're not discussing the value of
    the answer, but the SCATTER in that value.
    
    And that has sent me to an oldish textbook, M J Moroney's "Facts from
    figures" (1951). Which reminds me that in a Gaussian distribution, , though
    the standard error of the mean is sigma / root-n, the standard error of the
    median is 1.25 x sigma / root-n, where sigma is the standard deviation of
    the individual observations, and n is the number of observations.
    
    So, with a Gaussian distribution, if you take the median instead of the
    mean, then you will need to take 1.25-squared more observations to get as
    good an answer.
    
    But really, you'll need someone who knows much more about statistics that I
    do, to explain further.
    
    George.
    
    contact George Huxtable, at george{at}hux.me.uk
    or at +44 1865 820222 (from UK, 01865 820222)
    or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
    
    
    
    
    

       
    Reply
    Browse Files

    Drop Files

    NavList

    What is NavList?

    Join NavList

    Name:
    (please, no nicknames or handles)
    Email:
    Do you want to receive all group messages by email?
    Yes No

    You can also join by posting. Your first on-topic post automatically makes you a member.

    Posting Code

    Enter the email address associated with your NavList messages. Your posting code will be emailed to you immediately.
    Email:

    Email Settings

    Posting Code:

    Custom Index

    Subject:
    Author:
    Start date: (yyyymm dd)
    End date: (yyyymm dd)

    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site