In probability theory and statistics, correlation, also called correlation coefficient, is a numeric measure of the strength of linear relationship between two random variables. It is found by dividing their covariance by the product of their standard deviations, and was introduced by Francis Galton.

Mathematical properties

The correlation ρ_xy between two random variables X and Y with expected values μ_X and μ_Y and standard deviations σ_X and σ_Y is defined as:

Since μ_X=E(X), σ_X²=E(X²)-E²(X) and likewise for Y, we may also write:

The correlation is defined only if both standard deviations are finite and at least one of them is nonzero. It is a corollary of the Cauchy-Schwarz inequality that the correlation cannot exceed 1 in absolute value.

The correlation is 1 in the case of an increasing linear relationship, −1 in the case of a decreasing linear relationship, and some value in between in all other cases, indicating the degree of linear dependence between the variables. The closer the coefficient is to either −1 or 1, the stronger the correlation between the variables.

If the variables are independent then the correlation is 0, but the converse is not true because the correlation coefficient detects only linear dependencies between two variables. Here is an example: Suppose the random variable X is uniformly distributed on the interval from −1 to 1, and Y = X². Then Y is completely determined by X, so that X and Y are as far from being independent as two random variables can be, but their correlation is zero; they are uncorrelated. However, in the special case when X and Y are jointly normal, independence is equivalent to uncorrelatedness.

The sample correlation

If we have a series of n  measurements of X  and Y  written as x_i  and y_i  where i = 1, 2, ..., n, then the Pearson product-moment correlation coefficient can be used to estimate the correlation of X  and Y . The Pearson coefficient is also known as the "sample correlation coefficient". It is especially important if X  and Y  are both normally distributed. The Pearson correlation coefficient is then the best estimate of the correlation of X  and Y . The Pearson correlation coefficient is written:

where and are the sample means of x_i  and y_i , s_x  and s_y  are the sample standard deviations of x_i  and y_i  and the sum is from i = 1 to n. As with the population correlation, we may rewrite this as

Again, as is true with the population correlation, the absolute value of the sample correlation must be less than or equal to 1.

The sample correlation coefficient is the fraction of the variance in y_i  that is accounted for by a linear fit of x_i  to y_i . This is written

where σ_y|x²  is the square of the error of a linear fit of y_i  to x_i  by the equation y = a + bx.

and σ_y²  is just the variance of y

Note that since the sample correlation coefficient is symmetric in x_i  and y_i , we will get the same value for a fit of x_i  to y_i :

This equation also gives an intuitive idea of the correlation coefficient for higher dimensions. Just as the above described sample correlation coefficient is the fraction of variance accounted for by the fit of a 1-dimensional linear submanifold to a set of 2-dimensional vectors (x_i , y_i ), so we can define a correlation coefficient for a fit of an m-dimensional linear submanifold to a set of n-dimensional vectors. For example, if we fit a plane z = a + bx + cy  to a set of data (x_i , y_i , z_i ) then the correlation coefficient of z  to x  and y  is

"Correlation does not imply causation"

The conventional dictum that "correlation does not imply causation" is treated in the article titled spurious relationship. See also Correlation implies causation (logical fallacy). However, correlations have causes.

Non-parametric statistics

Pearson's correlation coefficient is a parametric statistic, and it may be less useful if the underlying assumption of normality is violated. Non-parametric correlation methods, such as Spearman's ρ and Kendall's τ may be useful when distributions are not normal; they are a little less powerful than parametric methods if the assumptions underlying the latter are met, but are less likely to give distorted results when the assumptions fail.

Other measures of dependence among random variables

To get a measure for more general dependencies in the data (also nonlinear) it is better to use the correlation ratio which is able to detect almost any functional dependency, or mutual information which detects even more general dependencies. To fully capture the dependence between random variables we must consider the copula between them.

External links

• Statsoft Electronic Textbook
• Pearson's Correlation Coefficient - How to calculate it fast