I am new to statistics and I understand that when a pattern can be found in the plots of the residuals and variables of a regression this is an indication that there is something which is missing from the regression.

I made a regression with two parameters, X1 and X2, for the regression I used: X1, X1^2, X2*X2, X2.

From this I got the following plots: enter image description here

So far, I think there might be an error with X1^2, because there seems to be a slight curvature in the plot. All the others look fairly spread out around the horizontal axis.

What could I deduce from such plots ?

Here is the Summary for more information:

enter image description here

  • $\begingroup$ Well, there are few data points so difficult to interpret. I think you may be overinterpreting, I would not be worried just because of this plots. $\endgroup$ – kjetil b halvorsen Sep 23 '16 at 17:19
  • $\begingroup$ Just had a meeting with a statistician at my university today. And he said the same thing, over interpretation and data points are too clustered. $\endgroup$ – Sorade Sep 23 '16 at 19:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.