+ Reply to Thread
Results 1 to 4 of 4

Thread: Leave one out cross validation

  1. #1
    Points: 1,765, Level: 24
    Level completed: 65%, Points required for next Level: 35

    Posts
    26
    Thanks
    20
    Thanked 0 Times in 0 Posts

    Leave one out cross validation




    Hi everybody,
    I've done a linear regression analysis with forward selection method. my number of observations are 32 and the final model selected 2 independents out of 27 variables entered to the regression analysis. now, I am asked to do a leave one out cross validation for checking the predictivity of the model. but I know a few about statistics and LOOCV. I wonder if it is possible to do LOOCV in SPSS and if it is possible, how should I do that.
    Regards
    Last edited by Bahareh; 09-29-2015 at 10:42 AM.

  2. #2
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Leave one out cross validation

    hi,
    I dont lnow SPSS but I suspect that it is already in the output somewhere. E.g. in Minitab it is called PRESS , predictive sum of squares. Maybe you will find something similar.

    regards

  3. The Following User Says Thank You to rogojel For This Useful Post:

    Bahareh (09-29-2015)

  4. #3
    Points: 1,765, Level: 24
    Level completed: 65%, Points required for next Level: 35

    Posts
    26
    Thanks
    20
    Thanked 0 Times in 0 Posts

    Re: Leave one out cross validation

    Thanks a lot. I saw leave one out cross validation option in Partial Least Square regression and PRESS value in Minitab. but is it possible to do cross validation for multiple linear regression in Minitab? and how should I interprete value of PRESS as an index of model validity?
    Regards

  5. #4
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Leave one out cross validation


    Hi,
    yes, you just need to check the option in Minitab, in the Outputs dialog box, if memory serves.

    Press actually means predicred sum if squares so the smaller the better.

    regards

  6. The Following User Says Thank You to rogojel For This Useful Post:

    Bahareh (10-07-2015)

+ Reply to Thread

           




Tags for this Thread

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