I have a doubt in some concepts in the subject of data warehouse , if someone could clarify it for me I would be very grateful.
To explain my doubt, I will present the following example. There is the following dimensional scheme :
Taking into account that there are measures of the Additive, Semi-Additive and Non-Additive type .
How do you insert the average data into the fact table since an average is a coarser-grained measure? (Different granularity).
First of all, Thanks.
The average cannot be aggregated in a fact table.
The average is not a measure that can be summarized, or calculated based on averages. It is a measure that only serves on the universe to take into account. Therefore, this measurement should always be calculated at the time of need.
If you had a table that said:
Note that if we wanted to calculate the total average, it is not enough to add the averages, but rather we have to add the quantity and the return prices:
Therefore, except as a point reference, it is not useful to have an average at any level of facts.