+ Reply to Thread
Results 1 to 3 of 3

Thread: backtranformation of log transformed data

  1. #1
    Points: 4,412, Level: 42
    Level completed: 31%, Points required for next Level: 138

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    backtranformation of log transformed data




    Hi forum,

    I've a simple question that has probably a difficult answer.

    In a lineair regression of the form:

    log(y) ~ x

    resulting in an equation log(y) = a + bx

    the backtransform would be the exp() of both sides.

    but how to handle the case:

    log(y) ~ x1 + x2 + x3

    how do you backtransform the resulting equation with multiple variables? It seems to work out in some cases but also produces strange outliers.

    Best regards,
    Koen

  2. #2
    Points: 33, Level: 1
    Level completed: 66%, Points required for next Level: 17

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: backtranformation of log transformed data

    I've looked all over for this answer myself and can never get the output to convert correctly in excel. I tried the =exp formula around the whole equation but the results looked spurious.

    In the end I went for a square root transformation instead as that's easy to understand . I'd love to see the formula to back transform a log lin model though so hopefully someone here will advise.

  3. #3
    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: backtranformation of log transformed data


    There is a 2015 article in "Epidemiology" that lists all of the backtransformations and is where the table comes from.
    Attached Images  
    Stop cowardice, ban guns!

+ Reply to Thread

           




Similar Threads

  1. Replies: 2
    Last Post: 05-25-2011, 12:36 AM
  2. Correlation Coefficient of Transformed Data
    By Tayyib in forum Statistics
    Replies: 0
    Last Post: 12-02-2009, 01:41 PM
  3. Replies: 1
    Last Post: 08-28-2008, 08:06 PM
  4. Log transformed statistics
    By kh11 in forum Statistics
    Replies: 3
    Last Post: 08-19-2008, 02:53 PM
  5. Do I need to center a transformed variable?
    By Mindy1979 in forum Statistics
    Replies: 1
    Last Post: 04-23-2008, 02:44 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