Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Mean Absolute Percentage Error in R

  1. Jul 24, 2013 #1
    Hi all,

    I'm trying to calculate the M.A.P.E. using the accuracy() function in SDMtools.

    I have an output of 10.69. I'm not sure if this is 10.69% or 1069% as previously in a different model I had gotten an output of 0.34.

    If anyone is familiar with this, I would appreciate any help.

    Thanks.
     
  2. jcsd
  3. Jul 24, 2013 #2

    mathman

    User Avatar
    Science Advisor
    Gold Member

    Without context, it is very hard to respond.
     
  4. Jul 24, 2013 #3
    Well I assumed the answer would be objective given that you are always required to enter the observed and predicted values.

    Regardless, what context would you like?
     
  5. Jul 25, 2013 #4

    mathman

    User Avatar
    Science Advisor
    Gold Member

    You are using something called SDMtools. I would hope that there would be a description of the units, etc. of its input and output. Without such a description, 10.69 has no meaning.
     
  6. Jul 25, 2013 #5

    MarneMath

    User Avatar
    Education Advisor

    I thought about this for a day, and it's been a long time since I've used R. However, using accuracy() does not return a MAPE. However, the package for time series and linear modeling, (I think it's simply called Forecast package, been awhile) does return a MAPE and if you are using that then your answer would be in a percentage.
     
  7. Jul 25, 2013 #6
    Yeah MarneMath, I think you're right. Thanks!
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Mean Absolute Percentage Error in R
  1. Mean square error (Replies: 2)

Loading...