Purpose
This function calculates the cumulative distribution of a value in a set of values. It is similar to the CUME_DIST function. You can use this function as an aggregate or analytic function.
As an aggregate function,
PERCENT_RANKcalculates the percentage position of a value in a set. The return value ranges from (0, 1]. If there areNrows of data, and the value ofexpris greater than the value of the second row and less than the value of the third row, the percentage position is2/N. Constant parameter expressions and expressions in theORDER BYclause of the aggregate function must match in position. Therefore, the number of parameters must be the same, and their types must be compatible. If a specified value is repeated in the grouping sequence, the two identical values are treated as one value.As an analytic function,
PERCENT_RANKcalculates the percentage rank for each row after sorting by one or more columns. The return value ranges from [0, 1]. When there are duplicate sorted values, the same rank is assigned to the rows with the same values, and the count of rows with the same values is recorded in the next rank. ThePERCENT_RANKvalue for the first row in any set is0. The formula for calculating the percentage position is: percentage position = rank / maximum rank. For example, see the table below:NRankPercentage PositionA 0 0 B 1 0.25 C 2 0.5 D 3 0.75 E 4 1
Syntax
/*Aggregate syntax*/
PERCENT_RANK(expr [, expr ...]) WITHIN GROUP
( ORDER BY expr_col [ DESC | ASC ][ NULLS { FIRST | LAST } ]
[,expr_col [ DESC | ASC ][ NULLS { FIRST | LAST } ]]...
)
/*Analytic syntax*/
PERCENT_RANK( ) OVER ([query_partition_clause] order_by_clause)
Parameters
Parameter |
Description |
|---|---|
| expr | Specifies the value to query. |
| expr_col | Specifies the column name corresponding to the value to query. |
| DESC | ASC | Specifies the sorting order. This is an optional parameter.
|
| NULLS { FIRST | LAST } | Specifies the position of NULL values in the expr_col after sorting. This is an optional parameter.
|
| OVER | Specifies the window to calculate using the OVER clause. For more information, see Analytic Function Description. |
Return Type
The return type is NUMBER.
Examples
Assume that the emp_msg table has been created.
obclient> SELECT * FROM emp_msg;
+--------+--------+------+------+
| DEPTNO | ENAME | SAL | MGR |
+--------+--------+------+------+
| 10 | CLARK | 2750 | 7839 |
| 10 | KING | 5300 | NULL |
| 10 | MILLER | 1600 | 7782 |
| 20 | ADAMS | 1400 | 7788 |
| 20 | FORD | 3300 | 7566 |
| 20 | JONES | 3275 | 7839 |
| 20 | SCOTT | 3300 | 7566 |
| 20 | SMITH | 1100 | 7902 |
| 30 | ALLEN | 1900 | 7698 |
| 30 | BLAKE | 3150 | 7839 |
| 30 | JAMES | 1250 | 7698 |
| 30 | MARTIN | 1550 | 7698 |
| 30 | TURNER | 1800 | 7698 |
| 30 | WARD | 1550 | 7698 |
| 30 | SCLARK | 1750 | 7839 |
+--------+--------+------+------+
15 rows in set
Aggregate Function Example
Return the percentage position of 3300 in the sal column.
obclient> SELECT PERCENT_RANK(3300) WITHIN GROUP (ORDER BY sal) FROM emp_msg;
+-------------------------------------------+
| PERCENT_RANK(3300)WITHINGROUP(ORDERBYSAL) |
+-------------------------------------------+
| .8 |
+-------------------------------------------+
1 row in set
Analytic Function Example
Group by the deptno column and sort the sal column in descending order, then return the percentage position of each value in the sal column.
obclient> SELECT deptno,ename,sal,
PERCENT_RANK() OVER (PARTITION BY deptno ORDER BY sal DESC) AS pr1
FROM emp_msg;
+--------+--------+------+-------------------------------------------+
| DEPTNO | ENAME | SAL | PR1 |
+--------+--------+------+-------------------------------------------+
| 10 | KING | 5300 | 0 |
| 10 | CLARK | 2750 | .5 |
| 10 | MILLER | 1600 | 1 |
| 20 | FORD | 3300 | 0 |
| 20 | SCOTT | 3300 | 0 |
| 20 | JONES | 3275 | .5 |
| 20 | ADAMS | 1400 | .75 |
| 20 | SMITH | 1100 | 1 |
| 30 | BLAKE | 3150 | 0 |
| 30 | ALLEN | 1900 | .1666666666666666666666666666666666666667 |
| 30 | TURNER | 1800 | .3333333333333333333333333333333333333333 |
| 30 | SCLARK | 1750 | .5 |
| 30 | MARTIN | 1550 | .6666666666666666666666666666666666666667 |
| 30 | WARD | 1550 | .6666666666666666666666666666666666666667 |
| 30 | JAMES | 1250 | 1 |
+--------+--------+------+-------------------------------------------+
15 rows in set