The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
Hi,
Share some data and show the expected result.
Here is a set of dummy data with a demo of what I am trying to do. The last column is what I am trying to create in BI, but I want to do this calculation using different filter contexts changing the # of "widgets sold" for non 1 year periods back through time:
Widgets | Dates | Rolling Average Previous 12 Months | Rolling Average Month 13-24 | Growth Between Two Rolling Averages |
146 | Jan-16 | N/a | N/a | N/a |
322 | Feb-16 | N/a | N/a | N/a |
131 | Mar-16 | N/a | N/a | N/a |
335 | Apr-16 | N/a | N/a | N/a |
258 | May-16 | N/a | N/a | N/a |
180 | Jun-16 | N/a | N/a | N/a |
110 | Jul-16 | N/a | N/a | N/a |
345 | Aug-16 | N/a | N/a | N/a |
385 | Sep-16 | N/a | N/a | N/a |
135 | Oct-16 | N/a | N/a | N/a |
250 | Nov-16 | 236.0909091 | N/a | N/a |
418 | Dec-16 | 251.25 | N/a | N/a |
267 | Jan-17 | 261.3333333 | N/a | N/a |
19 | Feb-17 | 236.0833333 | N/a | N/a |
300 | Mar-17 | 250.1666667 | N/a | N/a |
252 | Apr-17 | 243.25 | N/a | N/a |
264 | May-17 | 243.75 | N/a | N/a |
123 | Jun-17 | 239 | N/a | N/a |
94 | Jul-17 | 237.6666667 | N/a | N/a |
1 | Aug-17 | 209 | N/a | N/a |
294 | Sep-17 | 201.4166667 | N/a | N/a |
476 | Oct-17 | 229.8333333 | N/a | N/a |
373 | Nov-17 | 240.0833333 | 236.0909091 | 2% |
146 | Dec-17 | 217.4166667 | 251.25 | -13% |
385 | Jan-18 | 227.25 | 261.3333333 | -13% |
21 | Feb-18 | 227.4166667 | 236.0833333 | -4% |
11 | Mar-18 | 203.3333333 | 250.1666667 | -19% |
219 | Apr-18 | 200.5833333 | 243.25 | -18% |
88 | May-18 | 185.9166667 | 243.75 | -24% |
7 | Jun-18 | 176.25 | 239 | -26% |
297 | Jul-18 | 193.1666667 | 237.6666667 | -19% |
363 | Aug-18 | 223.3333333 | 209 | 7% |
18 | Sep-18 | 200.3333333 | 201.4166667 | -1% |
396 | Oct-18 | 193.6666667 | 229.8333333 | -16% |
242 | Nov-18 | 182.75 | 240.0833333 | -24% |
350 | Dec-18 | 199.75 | 217.4166667 | -8% |
341 | Jan-19 | 196.0833333 | 227.25 | -14% |
65 | Feb-19 | 199.75 | 227.4166667 | -12% |
491 | Mar-19 | 239.75 | 203.3333333 | 18% |
442 | Apr-19 | 258.3333333 | 200.5833333 | 29% |
8 | May-19 | 251.6666667 | 185.9166667 | 35% |
156 | Jun-19 | 264.0833333 | 176.25 | 50% |
306 | Jul-19 | 264.8333333 | 193.1666667 | 37% |
243 | Aug-19 | 254.8333333 | 223.3333333 | 14% |
Hi,
The third colum there seems like a measure. Can you share the raw data.
Hi datakid1230,
Remove the all statement and try again:
Result = CALCULATE ( [Daily Average Sales], DATESBETWEEN ( 'Calendar'[Date].[Date], DATE ( 2017, 8, 1 ), DATE ( 2018, 3, 31 ) ), DATESBETWEEN ( 'Calendar'[Date], LASTDATE ( 'Calendar'[Date] ) - 365, LASTDATE ( 'Calendar'[Date] ) ) )
Regards,
Jimmy Tao
Hi @Anonymous
Can you share the pbix? It would be much easier to see what is going on
No I can't share the information sorry. No worries if there is not enough information to provide an answer.