Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
Does anyone know how they calculate the percentiles on the analytics pane?
I added onto my graph the 90th percentile line from the anlaytics pane but it is giving me a different value to my calculated value (=percentile.inc(,0.9)).
This is the same when I use the average line and calculate the average (=average()).
Does anyone know why the values are different and how I can get them the same?
Thanks
P.s. I do have a few filters applied on the page but when I change the filters, the calculated fields and the analytic lines change with them so I don't think thats the problem.
Hi @Silver,
It's the context that has the magic. You can try a measure like this. I tested it in a simple column chart. The axis is 'Calendar[Date' while the value is SUM(Sales[Quantity]).
average171024 = CALCULATE ( AVERAGEX ( SUMMARIZE ( Sales, Calender[Date], "sq", SUM ( Sales[Quantity] ) ), [sq] ), ALL ( 'Calender' ) )
If you need more help, please provide a sample .pbix file.
Best Regards!
Dale
I do not want to come across as rude, but that answer makes no sense and does not address the orignal issue at all.
Best regards though!!
Thanks for the reply. I've tried your formula but it gave me a really large number.
Here is an example of my data:
Colour Letter Date/time value
red A 2017/12/12 01:00:00 2
blue A 2017/12/12 01:00:00 5
red A 2017/12/12 12:00:00 4
green B 2017/12/12 08:00:00 1
red A 2017/12/12 14:00:00 6
My line graph would be Average of value (y axis) against date/time (x -axis) with colour and letter as filters.
so technically, I want the percentiles for red,A.
User | Count |
---|---|
140 | |
113 | |
104 | |
77 | |
65 |
User | Count |
---|---|
136 | |
118 | |
101 | |
71 | |
61 |