Oracle Analytics and AI

BenFreez

Hi all,

i have a question on how to manage partial or full quarter.

From the example below, you can see that i have only value for July and August for Q3.

The idea is to divide by 3 the sum of the value when the Quarter is full and divide by the available month number that i have when the quarter is not full

For example, for Q3, Sum(160 + 100) / 2 because the quarter is partial.

Do you have an idea on how to implement this in the RPD ?

Thanks in adavance for your help.

Ben

2016-Q1	Feb-2016	105
	Jan-2016	230
	Mar-2016	70
2016-Q2	Apr-2016	65
	Jun-2016	145
	May-2016	80
2016-Q3	Aug-2016	100
	Jul-2016	160
	Sep-2016

mac2

I might be missing something here, but it sounds like you just want to do a straight up average of the data. So something like AVG(<your_fact_column> by <your_quarter_column>)

EDIT: Since you asked about doing in the RPD, and my last response is how you'd do it in an ad-hoc request... You probably already have a time dimension hierarchy modeled (if not, you need one first) and then you would AVG(<your_fact_column>) and set the level at the Quarter level.