Forum Discussion
Getting Previous Row using DAX measure
- 3 years ago
Hi,
Thank you for your sharing.
Could you please try the below if it works?
Prev_value_Energy_consum = VAR Index = [Index] VAR Prev_datetime = MAXX ( FILTER ( ALLSELECTED ( FloatTable ), FloatTable[Date] = SELECTEDVALUE ( FloatTable[Date] ) && FloatTable[Time] < SELECTEDVALUE ( FloatTable[Time] ) ), FloatTable[Time] ) VAR Prev_date = MAXX ( FILTER ( ALLSELECTED ( FloatTable ), FloatTable[Date] < SELECTEDVALUE ( FloatTable[Date] ) ), FloatTable[Date] ) VAR Prev_time = MAXX ( FILTER ( ALLSELECTED ( FloatTable ), FloatTable[Date] = Prev_date ), FloatTable[Time] ) RETURN SWITCH ( TRUE (), NOT ISBLANK ( Prev_datetime ), CALCULATE ( SUMX ( FloatTable, VAR Index = FloatTable[TagIndex] RETURN FloatTable[Val] ), FILTER ( ALLSELECTED ( FloatTable ), FloatTable[Time] = Prev_datetime && FloatTable[Date] = SELECTEDVALUE ( FloatTable[Date] ) && FloatTable[TagIndex] = Index ) ), CALCULATE ( SUMX ( FloatTable, VAR Index = FloatTable[TagIndex] RETURN FloatTable[Val] ), FILTER ( ALLSELECTED ( FloatTable ), FloatTable[Time] = Prev_time && FloatTable[Date] = Prev_date && FloatTable[TagIndex] = Index ) ) )
Hi Jihwan_Kim
Actually we have a column called 'TagIndex' which have values from 0 to 8000 and each Tag index refers to a pump located at different regions. So basically these are different pumps.
And There is a column called 'Val' which has cumulative value coming from database, we need to subtract current row from previous row to convert it to non cumulative value.
So that should be the main purpose of the logic, so I am creating a measure to get Prev_row for that record based on different tagindexes I select. after that I will create a calculated column which will subtract value from this measure.
I tried your formula but it is still showing those blank values at some point of time as I showed in screenshot.
Thanks and Regards
Mihir
Hi,
Thank you for your feedback.
Could you please share your sample pbix file's link (onedrive, googledrive, dropbox, any other) here, and then I can try to look into it to come up with a more accurate solution.
Thank you.