User Guide

10.2.2 HotFix 1
- 10.5.7
- 10.5.6
- 10.5.5
- 10.5.4
- 10.5.3
- 10.5.2
- 10.5.1
- 10.5
- 10.4.1
- 10.4.0
- 10.2.2 Service Pack 1
- 10.2.2
- 10.2.1

Back Next

Aggregate Functions as Window Functions

In addition to LEAD and LAG, you can also use aggregate functions as window functions. When you use aggregate functions like SUM and AVG as window functions, you can perform running calculations that are similar to the stateful functions MOVINGSUM, MOVINGAVG, and CUME. Window functions are more flexible than stateful functions because you can set a specific end offset.

To use an aggregate function as a window function, you must define a frame in the windowing properties. You define a frame to limit the scope of the calculation. The aggregate function performs a calculation across the frame and produces a single value for each row.

Example

You are a lumber salesperson who sold different quantities of wood over the past two years. You want to calculate a running total of sales quantities.

The following table lists each sale ID, the date, and the quantity sold:

Sale_ID	Date	Quantity
30001	2016-08-02	10
10001	2016-12-24	10
10005	2016-12-24	30
40001	2017-01-09	40
10006	2017-01-18	10
20001	2017-02-12	20

A SUM function adds all the values and returns one output value. To get a running total for each row, you can define a frame for the function boundaries.

The start offset is all rows preceding and the end offset is 0. The order key is ascending date. The partition key is not specified.

You configure the following windowing properties:

Start offset: All Rows Preceding

End offset: 0

Order Key: Date Ascending

Partition Key: Not specified

You define the following aggregate function:

SUM (Quantity)

SUM adds the quantity in the current row to the quantities in all the rows preceding the current row. The function returns a running total for each row.

The following table lists a running sum for each date: