In statistics, the Kendall rank correlation coefficient, commonly referred to as Kendall's ? coefficient (after the Greek letter ?, tau), is a statistic used to measure the ordinal association between two measured quantities. A ? test is a non-parametric hypothesis test for statistical dependence based on the ? coefficient.
It is a measure of rank correlation: the similarity of the orderings of the data when ranked by each of the quantities. It is named after Maurice Kendall, who developed it in 1938, though Gustav Fechner had proposed a similar measure in the context of time series in 1897.
Intuitively, the Kendall correlation between two variables will be high when observations have a similar (or identical for a correlation of 1) rank (i.e. relative position label of the observations within the variable: 1st, 2nd, 3rd, etc.) between the two variables, and low when observations have a dissimilar (or fully different for a correlation of -1) rank between the two variables.
Let be a set of observations of the joint random variables X and Y, such that all the values of () and () are unique (ties are neglected for simplicity). Any pair of observations and , where , are said to be concordant if the sort order of and agrees: that is, if either both and holds or both and ; otherwise they are said to be discordant.
The Kendall ? coefficient is defined as:
Where is the binomial coefficient for the number of ways to choose two items from n items.
The denominator is the total number of pair combinations, so the coefficient must be in the range -1 ?
The Kendall rank coefficient is often used as a test statistic in a statistical hypothesis test to establish whether two variables may be regarded as statistically dependent. This test is non-parametric, as it does not rely on any assumptions on the distributions of X or Y or the distribution of (X,Y).
Under the null hypothesis of independence of X and Y, the sampling distribution of ? has an expected value of zero. The precise distribution cannot be characterized in terms of common distributions, but may be calculated exactly for small samples; for larger samples, it is common to use an approximation to the normal distribution, with mean zero and variance
A pair is said to be tied if or ; a tied pair is neither concordant nor discordant. When tied pairs arise in the data, the coefficient may be modified in a number of ways to keep it in the range [-1, 1]:
where nc, nd and n0 are defined as in the next section.
The Tau-b statistic, unlike Tau-a, makes adjustments for ties. Values of Tau-b range from -1 (100% negative association, or perfect inversion) to +1 (100% positive association, or perfect agreement). A value of zero indicates the absence of association.
The Kendall Tau-b coefficient is defined as:
Be aware that some statistical packages, e.g. SPSS, use alternative formulas for computational efficiency, with double the 'usual' number of concordant and discordant pairs.
Tau-c (also called Stuart-Kendall Tau-c) is more suitable than Tau-b for the analysis of data based on non-square (i.e. rectangular) contingency tables. So use Tau-b if the underlying scale of both variables has the same number of possible values (before ranking) and Tau-c if they differ. For instance, one variable might be scored on a 5-point scale (very good, good, average, bad, very bad), whereas the other might be based on a finer 10-point scale.
The Kendall Tau-c coefficient is defined as:
When two quantities are statistically independent, the distribution of is not easily characterizable in terms of known distributions. However, for the following statistic, , is approximately distributed as a standard normal when the variables are statistically independent:
Thus, to test whether two variables are statistically dependent, one computes , and finds the cumulative probability for a standard normal distribution at . For a 2-tailed test, multiply that number by two to obtain the p-value. If the p-value is below a given significance level, one rejects the null hypothesis (at that significance level) that the quantities are statistically independent.
Numerous adjustments should be added to when accounting for ties. The following statistic, , has the same distribution as the distribution, and is again approximately equal to a standard normal distribution when the quantities are statistically independent:
This is sometimes referred to as the Mann-Kendall test.
The direct computation of the numerator , involves two nested iterations, as characterized by the following pseudocode:
numer := 0 for i := 2..N do for j := 1..(i - 1) do numer := numer + sign(x[i] - x[j]) × sign(y[i] - y[j]) return numer
Although quick to implement, this algorithm is in complexity and becomes very slow on large samples. A more sophisticated algorithm built upon the Merge Sort algorithm can be used to compute the numerator in time.
Begin by ordering your data points sorting by the first quantity, , and secondarily (among ties in ) by the second quantity, . With this initial ordering, is not sorted, and the core of the algorithm consists of computing how many steps a Bubble Sort would take to sort this initial . An enhanced Merge Sort algorithm, with complexity, can be applied to compute the number of swaps, , that would be required by a Bubble Sort to sort . Then the numerator for is computed as:
where is computed like and , but with respect to the joint ties in and .
A Merge Sort partitions the data to be sorted, into two roughly equal halves, and , then sorts each half recursive, and then merges the two sorted halves into a fully sorted vector. The number of Bubble Sort swaps is equal to:
where and are the sorted versions of and , and characterizes the Bubble Sort swap-equivalent for a merge operation. is computed as depicted in the following pseudo-code:
function M(L[1..n], R[1..m]) is i := 1 j := 1 nSwaps := 0 while i and j do if R[j] < L[i] then nSwaps := nSwaps + n - i + 1 j := j + 1 else i := i + 1 return nSwaps
A side effect of the above steps is that you end up with both a sorted version of and a sorted version of . With these, the factors and used to compute are easily obtained in a single linear-time pass through the sorted arrays.
cor.test(x, y, method = "kendall")in its "stats" package (also
cor(x, y, method = "kendall")will work, but without returning the p-value).