1

Correlation between two variables can be linear or non-linear. Accordingly different methods are used to find correlation coefficients.

My query is if the correlation can be linear or non- linear, why in most place/articles the definition of the correlation assume that the association is linear?

SR1
  • 31
  • 1
    Pearson correlation measures linear association. Kendall and Spearman and quadrant correlations measure monotonic association. When people say "correlation" without further qualification they usually intend Pearson and most people use Pearson, so most things that talk about correlation without any specific adjective will talk about linear correlation. – Glen_b Jul 27 '22 at 04:11
  • @Glen_b That looks like an answer to me! – Dave Jul 27 '22 at 04:13
  • 1
    Correlation is neither "linear" nor "non-linear," but there is a relationship between Pearson correlation and certain kinds of linear models. – whuber Jul 27 '22 at 15:12
  • Does that mean linear and non linera relationships are in context of variables not correlation? Correlation determines the strength between two variables. Method of testing correlation differs depending on the type of relationship between variables (linear vs non linear ) – SR1 Jul 27 '22 at 17:45

0 Answers0