Forum Discussion
Weighted Average in a Hierarchy Visualized by a Matrix
I am trying to create a measure to show the weighted average at multiple levels of a hiearchy. I will be presenting this data in a matrix visualization in Power BI. I will provide sample data to illustrate this in more detail.
In this example, I have three levels of the hierachy: Product; Location; and Business.
I want the average Unit Price weighted by Qty Sold.
In Excel, the results would look like this for Location:
and like this for Business:
In Power BI, the expectation is that when you collapse a level of the hierarchy in the matrix, you see the corresponding weighted average in an adjacent column. At the lowest level you would simply show the original Unit Price per product.
Is this possible? Is there another solution I should consider?
Try this:
Avg = DIVIDE ( SUMX ( Table1, Table1[Unit Price] * Table1[Qty Sold] ), SUMX ( Table1, Table1[Qty Sold] ) )
4 Replies
- AlexisOlson
Super User
Try this:
Avg = DIVIDE ( SUMX ( Table1, Table1[Unit Price] * Table1[Qty Sold] ), SUMX ( Table1, Table1[Qty Sold] ) )Thank you! This is perfect! However, I found one anomaly during my validation--maybe you will know what is causing it. One of the values at the lowest granularity came up as a lower value compared to the actual. But others seemed unafflicted by this. In this example I get ~$87k where I would expect ~$155k.
- AlexisOlson
Is your Unit Price column using a SUM aggregation? I bet you have more than one row in your data table for 1A and it's adding the prices together like it does for the subtotals.
