1

When I try to fit an intercept only model in R, why do I need get an R squared? I can see from the formula of the adjusted R squared that this would be 0, but this isn't shown either.

Bill
  • 556
  • 1
  • 9
  • 1
    This is more of a software question than a stats question, but I imagine the developers realized R squared is 0 when you don't have any predictors and so maybe thought "no use in showing it". – Demetri Pananos Oct 24 '20 at 17:11
  • 1
    The variability of a response is not explained by fitting a constant -- because that doesn't vary. – Nick Cox Oct 24 '20 at 18:53

0 Answers0