“Kendall rank correlation coefficient”的意思、由来-开放百科全书

In statistics, the Kendall rank correlation coefficient, commonly referred to as Kendall's tau coefficient (after the Greek letter τ), is a statistic used to measure the ordinal association between two measured quantities. A tau test is a non-parametric hypothesis test for statistical dependence based on the tau 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,^[1] though Gustav Fechner had proposed a similar measure in the context of time series in 1897.^[2]

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.

Both Kendall's

and Spearman's

can be formulated as special cases of a more general correlation coefficient.

Definition

Let (x₁, y₁), (x₂, y₂), ..., (x_n, y_n) be a set of observations of the joint random variables X and Y respectively, such that all the values of (

) and (

) are unique. Any pair of observations

and

, where

, are said to be concordant if the ranks for both elements (more precisely, the sort order by x and by y) agree: that is, if both

and

; or if both

and

. They are said to be discordant, if

and

; or if

and

. If

, the pair is neither concordant nor discordant.

Properties

The denominator is the total number of pair combinations, so the coefficient must be in the range −1 ≤ τ ≤ 1.

Hypothesis test

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 τ 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

Accounting for ties

A pair

is said to be tied if

; 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]:

Tau-a

The Tau-a statistic tests the strength of association of the cross tabulations. Both variables have to be ordinal. Tau-a will not make any adjustment for ties. It is defined as:

Tau-b

The Tau-b statistic, unlike Tau-a, makes adjustments for ties.^[5] 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.

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.^[6]

Tau-c

Tau-c (also called Stuart-Kendall Tau-c)^[7] is more suitable than Tau-b for the analysis of data based on non-square (i.e. rectangular) contingency tables.^[7]^[8] 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.

Significance tests

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:

Algorithms

The direct computation of the numerator

, involves two nested iterations, as characterized by the following pseudo-code:

Although quick to implement, this algorithm is

in complexity and becomes very slow on large samples. A more sophisticated algorithm^[9] 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:

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:

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.

See also

References

1. ^{{cite journal |last=Kendall |first=M. |year=1938 |title=A New Measure of Rank Correlation |journal=Biometrika |volume=30 |issue=1–2 |pages=81–89 |doi=10.1093/biomet/30.1-2.81 |jstor = 2332226}}
2. ^{{cite journal |last=Kruskal |first=W.H. |authorlink=William Kruskal |year=1958 |title=Ordinal Measures of Association |journal=Journal of the American Statistical Association |volume=53 |issue=284 |pages=814–861 |mr=100941 | jstor = 2281954 |doi=10.2307/2281954}}
3. ^{{SpringerEOM |title=Kendall tau metric |id=K/k130020 |first=R.B. |last=Nelsen}}
4. ^{{SpringerEOM |title=Kendall coefficient of rank correlation |id=K/k055200 |first=A.V. |last=Prokhorov}}
5. ^{{cite book |last=Agresti |first=A. |year=2010 |title=Analysis of Ordinal Categorical Data |edition=Second |location=New York |publisher=John Wiley & Sons |isbn=978-0-470-08289-8 }}
6. ^{{cite book|last1=IBM|title=IBM SPSS Statistics 24 Algorithms|date=2016|publisher=IBM|page=168|url=http://www-01.ibm.com/support/docview.wss?uid=swg27047033#en|accessdate=31 August 2017}}
7. ^¹{{cite journal |last1=Berry |first1=K. J. |last2=Johnston |first2=J. E. |last3=Zahran |first3=S. |last4=Mielke |first4=P. W. |year=2009 |title=Stuart's tau measure of effect size for ordinal variables: Some methodological considerations |journal=Behavior Research Methods |volume=41 |issue=4 |pages=1144–1148 |doi=10.3758/brm.41.4.1144 |pmid=19897822}}
8. ^¹{{cite journal |last=Stuart |first=A. |year=1953 |title=The Estimation and Comparison of Strengths of Association in Contingency Tables |journal=Biometrika |volume=40 |issue=1–2 |pages=105–110 |jstor = 2333101 |doi=10.2307/2333101 }}
9. ^{{cite journal |doi=10.2307/2282833 |last=Knight |first=W. |year=1966 |title=A Computer Method for Calculating Kendall's Tau with Ungrouped Data |journal=Journal of the American Statistical Association |volume=61 |issue=314 |pages=436–439 | jstor = 2282833}}