The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
I am creating a chart that compares an average value against a 1-hour moving average value. I have defined my moving average measure as:
1-Hour Rolling Average =
VAR _end =
max('Query Table'[start_time])
RETURN
CALCULATE(AVERAGE('Query Table'[average_queries]),
FILTER(
ALLSELECTED('Query Table'),
'Query Table'[start_time] <= _end
&& DATEDIFF('Query Table'[starttime], _end, MINUTE) <= 60
))
When I create my chart with my start_time, my average_queries field, and my 1-Hour Moving Average measure, though, I get the same moving average line for all small multiples:
I know that I need to do something to adjust the context the measure is operating in, but I'm stuck! Any tips would be appreciated.
Solved! Go to Solution.
Hi @willchamp ,
I think you can try to use ALLEXCEPT() function in your measure. You need to add the column in small multiple in it.
1-Hour Rolling Average =
VAR _end =
MAX ( 'Query Table'[start_time] )
RETURN
CALCULATE (
AVERAGE ( 'Query Table'[average_queries] ),
FILTER (
ALLEXCEPT ( 'Query Table', 'Query Table'[Column in Small Multiple] ),
'Query Table'[start_time] <= _end
&& DATEDIFF ( 'Query Table'[starttime], _end, MINUTE ) <= 60
)
)
Best Regards,
Rico Zhou
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi @willchamp ,
I think you can try to use ALLEXCEPT() function in your measure. You need to add the column in small multiple in it.
1-Hour Rolling Average =
VAR _end =
MAX ( 'Query Table'[start_time] )
RETURN
CALCULATE (
AVERAGE ( 'Query Table'[average_queries] ),
FILTER (
ALLEXCEPT ( 'Query Table', 'Query Table'[Column in Small Multiple] ),
'Query Table'[start_time] <= _end
&& DATEDIFF ( 'Query Table'[starttime], _end, MINUTE ) <= 60
)
)
Best Regards,
Rico Zhou
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
@willchamp , Use a datetime table join with datetime of your table. That will allow you not to use ALLSELECTED('Query Table') and other contexts may pass
Datetime = distinct('Query Table'[starttime])
1-Hour Rolling Average =
VAR _end =
max('DateTime'[starttime])
RETURN
CALCULATE(AVERAGE('Query Table'[average_queries]),
FILTER(
ALLSELECTED('DateTime'),
'DateTime'[starttime] <= _end
&& DATEDIFF('DateTime'[starttime], _end, MINUTE) <= 60
))
Or consider window function