0

Is the model below an example of nonlinear regression? $$y = \beta_1 + \beta_2^2x$$ Some books say, it is, but I'm not convinced as $\beta_2^2$ can itself considered a constant denoted by (say) $\beta_3$.

I have studied other examples of nonlinear regression and been convinced about them. The question differs from an answer suggested in that in my question, power is over $\beta$, the regression coefficient, while on the answer, power is over x, the independent variable.

Abdullah Al Mahmud
  • 11
  • 1
  • 4
  • 1
    Your question is directly and explicitly answered at point (5) in [my answer in the duplicate thread](https://stats.stackexchange.com/a/148713/919). – whuber Sep 24 '17 at 17:13
  • to me, it's not so 'explicit'. – Abdullah Al Mahmud Sep 24 '17 at 17:22
  • 1
    We have to draw the line somewhere. We don't need infinitely many questions dealing with problems that are simple variations of each other. If the existing answer is not clear to you, then please edit your question to indicate where you need help relating that answer to your problem. – whuber Sep 24 '17 at 17:27
  • 1
    You are right that $\beta_2^2$ can simply be written as $\beta_3$. Thus, there isn't anything much else to say here. For a comprehensive understanding of the issues pertaining to the different senses of 'nonlinear', you couldn't do better than to learn @whuber's linked answer. If there is a specific element of that that confuses you, you should quote that element, state what you understand about it, & what is confusing about it for you. – gung - Reinstate Monica Sep 25 '17 at 01:03

0 Answers0