Source: https://dev-docs.nebula-graph.io/v5.3.2/database-gql-reference/fe/expressions/aggregate/
Documentation version: v5.3.2

# Aggregate value expression

The aggregate value expression holds an aggregate function instance and returns the execution result of the aggregate function.

Aggregate value expressions can be used together with [GROUP BY clauses](https://dev-docs.nebula-graph.io/v5.3.2/database-gql-reference/dql/clauses/group-by/) in RETURN statements to group the result set by one or more columns. For more information, see [RETURN statement](https://dev-docs.nebula-graph.io/v5.3.2/database-gql-reference/dql/return/).

## Syntax

diagramsource

 aggregate_value_expression  

````
aggregate_value_expression ::=
    aggregate_function '(' ( 'DISTINCT' | 'ALL' )? value_expression ( ',' value_expression )* ')' 

````

## Parameters

### `aggregate_function`

The name of an aggregate function. For more information about aggregate functions, see [Aggregate functions](https://dev-docs.nebula-graph.io/v5.3.2/database-gql-reference/fe/functions/aggregate/).

### `DISTINCT | ALL`

`DISTINCT` and `ALL` are specified to indicate whether duplicate values are included in the specified group.

- If you specify `DISTINCT`, the aggregate function ignores duplicate values in the specified group.
- If you specify `ALL`, the aggregate function includes duplicate values in the specified group.
- If you do not specify `DISTINCT` or `ALL`, the default is `ALL`.

### `value_expression`

A value that is taken as an argument by the aggregate function. The value can be of any data type. For more information, see [Value expressions](https://dev-docs.nebula-graph.io/v5.3.2/database-gql-reference/fe/#value_expressions).

## Examples

The following examples are based on the [example dataset](https://dev-docs.nebula-graph.io/v5.3.2/database-gql-reference/overview/example-dataset/).

### Aggregation operations on a single group of values

````
FOR i in LIST [1,2,3,3,4,4] RETURN count(DISTINCT i), avg(i), sum(i), collect(DISTINCT i) GROUP BY ()

````

The preceding statement uses `GROUP BY ()` to group the entire result set as a single group and uses the `count`, `avg`, `sum`, and `collect` aggregate functions to perform aggregation operations on the group of rows. The `count(DISTINCT i)` expression returns the count of distinct values in the group. The `avg(i)` expression, equivalent to `avg(ALL i)`, returns the average of all values in the group. The `sum(i)` expression, equivalent to `sum(ALL i)`, returns the sum of all values in the group. The `collect(DISTINCT i)` expression returns a list consisting of the distinct values in the group.

Return:

````
+-------------------+--------------------+--------+---------------------+
| count(DISTINCT i) | avg(i)             | sum(i) | collect(DISTINCT i) |
+-------------------+--------------------+--------+---------------------+
| 4                 | 2.8333333333333335 | 17     | [4,3,2,1]           |
+-------------------+--------------------+--------+---------------------+

````

### Aggregation operations on the result of a MATCH statement

````
'''
USE movie 
MATCH ()-[e:Watch]->(v@Movie {id: 472}) 
RETURN collect(e.rate), avg(e.rate) GROUP BY ()
'''

````

The preceding statement returns all the ratings of the movie with ID `472` and calculates the average rating of the movie.

Return:

````
+-----------------+-------------+
| collect(e.rate) | avg(e.rate) |
+-----------------+-------------+
| [3.0]           | 3.0         |
+-----------------+-------------+

````