Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
Hi!
I have a dataset with accumulated values per random weeks.
I want to calculate hours per week. In the cases where a consecutive week is missing I would like to add the week and a zero value for each id.
Id | yyww | acc hours |
a | 2208 | 110 |
a | 2207 | 106 |
a | 2203 | 85 |
a | 2201 | 56 |
a | 2151 | 36 |
a | 2150 | 15 |
a | 2148 | 10 |
a | 2146 | 5 |
b | 2208 | 164 |
b | 2207 | 152 |
b | 2206 | 148 |
b | 2205 | 108 |
b | 2204 | 57 |
b | 2203 | 56 |
b | 2202 | 22 |
c | 2205 | 69 |
c | 2201 | 15 |
c | 2110 | 14 |
c | 2109 | 8 |
c | 2105 | 6 |
c | 2104 | 4 |
c | 2101 | 2 |
Any ideas?
/Anna.
Solved! Go to Solution.
Hi,
I suggest creating a calendar table like the attached pbix file, and create a measure like below.
Calendar Table:
Calendar =
ADDCOLUMNS (
CALENDAR ( DATE ( 2021, 1, 1 ), DATE ( 2022, 12, 31 ) ),
"YYYYWW",
YEAR ( [Date] ) * 100
+ WEEKNUM ( [Date], 21 )
)
Measure:
Hours: =
VAR currentweeknumber =
MAX ( 'Calendar'[YYYYWW] )
VAR acc_hours =
CALCULATE (
SUM ( Data[acc hours] ),
FILTER ( Data, Data[yyww] + 200000 = currentweeknumber )
)
VAR previousweeknumber =
MAXX ( FILTER ( Data, Data[yyww] + 200000 < currentweeknumber ), Data[yyww] ) + 200000
VAR previousacchours =
CALCULATE (
SUM ( Data[acc hours] ),
FILTER ( Data, Data[yyww] + 200000 = previousweeknumber )
)
RETURN
IF ( acc_hours <> 0, acc_hours - previousacchours, 0 )
If this post helps, then please consider accepting it as the solution to help other members find it faster, and give a big thumbs up.
Hi,
I suggest creating a calendar table like the attached pbix file, and create a measure like below.
Calendar Table:
Calendar =
ADDCOLUMNS (
CALENDAR ( DATE ( 2021, 1, 1 ), DATE ( 2022, 12, 31 ) ),
"YYYYWW",
YEAR ( [Date] ) * 100
+ WEEKNUM ( [Date], 21 )
)
Measure:
Hours: =
VAR currentweeknumber =
MAX ( 'Calendar'[YYYYWW] )
VAR acc_hours =
CALCULATE (
SUM ( Data[acc hours] ),
FILTER ( Data, Data[yyww] + 200000 = currentweeknumber )
)
VAR previousweeknumber =
MAXX ( FILTER ( Data, Data[yyww] + 200000 < currentweeknumber ), Data[yyww] ) + 200000
VAR previousacchours =
CALCULATE (
SUM ( Data[acc hours] ),
FILTER ( Data, Data[yyww] + 200000 = previousweeknumber )
)
RETURN
IF ( acc_hours <> 0, acc_hours - previousacchours, 0 )
If this post helps, then please consider accepting it as the solution to help other members find it faster, and give a big thumbs up.
Hi! This certainly seems to work for you but I can't get the total per week right. I think I have managed using "Earlier" and a group index I found here https://radacad.com/create-row-number-for-each-group-in-power-bi-using-power-query
Hi @Anonymous ,
Please use date table for week calculation and return zero for blank values.
Thanks,
Samarth
Best Regards,
Samarth
If this post helps, please consider accepting it as the solution to help the other members find it more quickly.
Appreciate your Kudos!!
Connect on Linkedin
User | Count |
---|---|
124 | |
108 | |
99 | |
62 | |
62 |
User | Count |
---|---|
137 | |
115 | |
102 | |
71 | |
61 |