It's known that Pearson's correlation is able to measure trends for an observed sample $y$ and a possible linear relationship with a simulated data $y^{(s)}$, being $+1$ if $y^{(s)} = a + by$ and $b >0$.

But, does it matter the shape of the observed data $y$? For instance, if I have $y = f(x) = x^2$, would it make any effect for Pearson's correlation? My intuition says no, but I want to find a more reliable source.

  • $\begingroup$ Are you alluding to the canonical example of zero correlation but a clear relationship: $E[X] = 0$, density of $X$ is symmetric, and $Y = X^2$? Then $Corr(X,Y) = 0$ but there is an entirely deterministic relationship. $\endgroup$ Jun 22, 2016 at 6:23
  • $\begingroup$ I think my question is much simpler than that. Does the characteristics of the observed sample matter for the Pearson? Does it have to obey any sort of characteristics? $\endgroup$
    – rph
    Jun 23, 2016 at 13:02
  • $\begingroup$ That has a simple mathematical answer, obtainable from any formula for the correlation: it is defined provided both variables have nonzero variance. $\endgroup$
    – whuber
    Sep 7, 2022 at 12:57
  • $\begingroup$ "Does the characteristics of the observed sample matter for the Pearson?" what does that term 'matter' actually mean in this context?. You can compute a Pearson correlation for any set of paired numbers, whether it is a linear relationship or not, so I would say that it doesn't matter. $\endgroup$ Sep 7, 2022 at 14:11

2 Answers 2


Pearson's correlation coefficient is a measure of strength of linear relationship between the variable. So, it may provide false results for non-linear relationship.

Read a more detailed answer on Correlation and dependence

  • 2
    $\begingroup$ I would say something like, "It may not pickup any aspect of a non-linear relationship." For example, if $X \sim N(0, 1)$ and $Y = X^2$, then $Corr(X,Y)=0$. That result is true! Saying it provides "false results" in this case isn't quite right. $\endgroup$ Jun 23, 2016 at 16:06

Adding on Dr Nisha Arora's answer, the following provides a method to compute meaningful Pearson linear correlation coefficient (PLCC) in such cases.

Suppose, you have $$y=f(x)=x^2$$ Take log on both sides $$log(y) = 2 \ log(x)$$ Now have a linear relationship. Find Pearson's Correlation Coefficient now. Read this answer for more details.

  • 2
    $\begingroup$ nonlinear transformations e.g. log-transformation will generally affect the correlation between 2 pairs. Reference $\endgroup$
    – Mario
    Sep 7, 2022 at 1:00
  • $\begingroup$ @Mario, of course. You're right. I've updated my answer. To further clarify, we are banking on that. Pearson correlation (PLCC) is defined for linear dependencies. For non-linear dependencies, PLCC does not make much sense. So, we convert our non-linear relationship to a linear one and then compute PLCC. This is also common in the literature. $\endgroup$ Sep 7, 2022 at 12:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.