# kendall

## Syntax

`kendall(X, Y)`

## Arguments

**X** is a scalar, vector, matrix or in-memory table.

**Y** is a scalar, vector, matrix or in-memory table.

## Details

Calculate the Kendall rank correlation coefficient
between *X* and *Y*. NULL values are ignored in the calculation.

If *X* or *Y* is a matrix, perform the aforementioned calculation on each
column and return a vector.

If *X* or *Y* is an in-memory table, perform the aforementioned calculation
on each numeric column of the table and return a table (where NULLs are returned for
non-numeric columns).

## Examples

```
x = [33,21,46,-11,78,47,18,20,-5,66]
y = [1,NULL,10,6,10,3,NULL,NULL,5,3]
kendall(x, y)
// output
0.05
```

If *X* is a matrix, *Y* can be a vector of the same length as the number of
rows in *X*, or a matrix of the same dimension as *X*. The result is a
vector of the same length as the number of columns in *X*.

```
m=1..20$10:2
kendall(x,m)
// output
[-0.0222,-0.0222]
n=rand(20,20)$10:2
kendall(m,n)
// output
[0.3865,-0.1591]
```

If *X* is a table, *Y* can be a vector of the same length as the number of
rows in *X*, or a table of the same dimension as *X*. The result is a
vector of the same length as the number of columns in *X*.

```
t=table(2..11 as id, "a"+string(2..11) as name)
kendall(t,x)
// output
id name
-0.0222
t1=table(x as col1, y as col2)
kendall(t,t1)
// output
id name
-0.0222
```