Hi guys,
I have a usecase where I need to compare Actual and Target values for a timeframe. There are several pitfalls included but one of them is that the granularity between the actual data table and the target data table (2 different fact tables) is different.
The actual table contains info on a daily basis across multiple years. The target table only contains the end values for 2023.
The expectation is to now somehow distribute the target value for the year 2023 so that a (probably at least monthly) comparison between actual and target is possible. They want a cumulative line chart as a result across the year to compare the actuals and targets.
The target values need to be whole numers (there are no 0.x targets). At the same time the targets need to be somehow linear distributed so you get a linear (in the chart cumulative) uptrend to the end of the year (final target)
Hi,
What do you mean by “The target table only contains the end values for 2023”? Is that month-end values or year-end (i.e. one value)? Do you have a data sample so we can understand better? My idea would be along the lines of joining the two tables based on a full date (dd/mm/yyyy) by adding a fixed day to the less granular data, then outer join the two tables and use the Missing Value node with the Linear Interpolation feature to create values for the missing days in the less granular column.
Kind regards,
Alexander
If i could understand, you’d like to create a projection/forecast with the previous data with dates and numbers, for then add the perspective of the future OR compare 2 tables into the same graphic lines, like the real numbers and the expectative from the same period?
Sorry if I made some mistakes about your question ok?
Yes it’s year end value
Quick and dirty:
I guess ich divide by 12 round up and reduce the final month value so it fits the target e.g.
target 100
100/12 = 8,x
Jan to Nov 9 and Dec 1