Forum Discussion
Resolver ICombining 2 Consecutive Measures (Historical and Projected)
The issue was that when the second measure uses the first one as a baseline in a separate table (with dates not connected), it calculates the final value for the first date and then uses that constant as the base for the rest of the measure. However, once you relate the dates to a table to combine them, I no longer get that final constant, and get a blank, as the first measure doesn't calculate as anything for the dates of the second measure.
I fixed this by keeping the historical measure as it was, and changing the projected measure to cover both the historical and projected portion. Then putting the historical as the line on top on the graph, so the only visible portion of the projected line is the actual projections.
Super UserHI askelton,
I made a sample of your data an tried the solution I gave you and as you can see the result is as expected, the only thing I cannot copy is the measures since I don't have the full layout of you setup, and did it with the values you placed in your images, however I believe that the result may be also achievable in your part.
Assumptions on my setup:
- 2 Tables - 1 Historical - 1 Projections
- 1 Calendar - This is made from a combination of unique values from Date column of the two previous tables
- Relationship One to Many
- Calendar[DateFixed] (One) -> Historical[Date] (Many)
- Calendar[DateFixed] (One) -> Projections[Date] (Many)
- Don't make filter for both sides
- Axis of chart is the Calendar[DateFixed] column
- Both Measures are placed in the lines
How do you have your calendar table made? How is it related with thew other tables?
I believe that you are getting this type of chart because of the way the calendar and the two other tables are related.
Regards,
MFelix
Resolver IThe issue was that when the second measure uses the first one as a baseline in a separate table (with dates not connected), it calculates the final value for the first date and then uses that constant as the base for the rest of the measure. However, once you relate the dates to a table to combine them, I no longer get that final constant, and get a blank, as the first measure doesn't calculate as anything for the dates of the second measure.
I fixed this by keeping the historical measure as it was, and changing the projected measure to cover both the historical and projected portion. Then putting the historical as the line on top on the graph, so the only visible portion of the projected line is the actual projections.
- MFelix
how does the tables relate with the calendar table?
'and changing the projected measure to cover both the historical and projected portion.'
PLease explain how you did that measure? I need as well such a dynamix measure including both - historical values up to yesterday - last historic valuw, and projected values up to the end of month - and it's replacing each day - when fact value appears.
