0

As far as I know, variables with p>0.05 are non-significant for the regression model.

I found similar questions here about p-value like: What is the meaning of p values and t values in statistical tests?

and this one Interpreting non-significant regression coefficients

But those don't answer my main question which is: can I exclude the variables with p>0.05 and run the model again until I get only significant variables?

David Nichols
  • 840
  • 2
  • 9
Hrpereira
  • 13
  • 3
  • 2
    Sure you can. But the result is rarely worth the effort and often is meaningless. See our higher-voted threads on variable selection and model building. – whuber Feb 17 '20 at 17:27
  • 2
    In line with whuber, the question whether you can and the question whether it's any good are quite different. ;-) – Christian Hennig Oct 31 '20 at 11:00
  • I see! Thanks whuber and Lewian So, although having non-significant variables, a model can be considered acceptable to define a specific case. It's a bit contra intuitive because it will increase the R^2, leadin Is that correct? Perhaps I need to create another question for this. – Hrpereira Nov 02 '20 at 15:09

0 Answers0