Group by in SAP PA 2.5 Data Manager

[Former Member]

Hello,

I am using SAP Predictive Analytics 2.5 and using data manager to create a data manipulation. I want to use create a new table from an existing table by summing (using aggregate function) the individual columns and grouping them based on two / three columns. So basically I'm trying to create a new table doing the function of below SQL query.

select f0, f1, sum(f2), sum(f3), sum(f4), count(f5) from table1 group by f0, f1

Thanks for your help!

[abdel_dadouche · Accepted solution]

Hi,

We use to have a feature call "BasicAggregates" in the old KXEN days, but this feature is not fully exposed and therefore supported.

What I can recommend is to submit a new entry in the Idea place for Predictive https://ideas.sap.com/PredictiveAnalytics. I will vote for that!

I will check with our Product Manager if this is still there and if I can share the usage of this feature.

Regards

@bdel

[marc_daniau]

You can create a new expression with formula: AggregateSum()

Then uncheck the Visibility check-box for the fields that you want to put in the group by

This function belongs to a family of basic aggregate functions: AggregateAvg() AggregateCount() AggregateMin() AggregateMax() AggregateSum()

They are not exposed in the user interface. I need to check if there is a setting to make them visible in the function list.

[marc_daniau]

Refer to the following post to see how to expose the basic aggregates functions.

http://scn.sap.com/message/16713096#16713096