Level-of-detail calculations known as LODs allow us to control the granularity of our data.
They have the format {Keyword [Dimesion X ] , [Dimesion Y ] : Aggregate expression }
Keyword = Fixed, Include, Exclude
An example of this would be looking at the sum of sales by Year/Month/Week/Day. We do this by "Fixing " the desired date "unit".
Fixed [ Year], [Region]: SUM (Sales) }
The above LOD would return the Sum of sales by region by year.
In the example below we have fixed Customer ID and Region. So we will be given the Minimum order date (the first one ) for each customer in each region.
If we remove the Customer ID we will be returned First Order by region.
We could of course replace Min with Max to get the highest or most recent date, as seen below again by Customer and Region.
