iSixSigma

R Square Prediction Question

Six Sigma – iSixSigma Forums Old Forums General R Square Prediction Question

This topic contains 1 reply, has 2 voices, and was last updated by  Enrico 11 years, 1 month ago.

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #50087

    lin
    Participant

    Hello,
    I ran a regression that gave me a fairly high R Squared (90%).  A couple of the factors have a high VIF.  But what puzzles me most is that  the R Square Prediction is 0%.  What causes this?  Thanks.

    0
    #172331

    Enrico
    Participant

    It’s quite easy… simply speaking the dataset you got has a lot of multicollinearity inside… multicollinearity is a near linear dependence among the regression variables (correlation between regressors)..
    look at the PRESS indicator… if this is high it means that there’s one or more high influence points, special causes inside dataset..
    As a solution to this the only way is to run a doe… “no causation without experimentation”
    regards
    Enrico

    0
Viewing 2 posts - 1 through 2 (of 2 total)

The forum ‘General’ is closed to new topics and replies.