Hi,
possibly the wrong forum, but seemed most appropriate:
currently have a table calculating the linest_m value for the last 12 weeks for a dimension selected via an island table, unfortunately having some trouble getting the Totals on the measure to reflect the linest_m of the data as a whole.
current expression is:
-LinEst_M({<WeekDiff={"<12"}>} aggr(sum({<WeekDiff={"<12"}>}ROWCOUNT),WeekDiff,[$(=DimIsland)]),WeekDiff)
this works at a row level for all dimensions however the totals dont work how id want.
ideally I'd want it to equal the below no matter what dimension it is split to
-LinEst_M({<WeekDiff={"<12"}>} aggr(sum({<WeekDiff={"<12"}>}ROWCOUNT),WeekDiff),WeekDiff)
If i change the totals function to sum rather than auto it is accurate for some dimensions (dimensions that are always there) but not for others (i think this is because for some dimensions there might not be a complete 12 week picture so the slope is being calculated over a smaller period - eg a staff member that started 8 weeks ago).
WeekDiff an integer counting down to zero ie 3 is 3 weeks ago, 0 is this week , hence the - at the start of the expressions
DimIsland is a field in an island table with the selectable Dimensions in.
Realise what i want might not be possible and i have currently got some KPI objects doing the totals but would be nice to incorporate into the table if at all possible.
If anyone has any ideas that would be great
