Forecasting is not summing up

Hi pfuschi,

If you mean look up suitable records based on year and age, you can refer to below steps.

1. Create a table stored year and not has relationship to original table.

Year Range = GENERATESERIES(2010,2040,1) 

2. Create matrix visual with age as row(original table), year as column(year range table), lookupvalue measure as value. 

Look up =
VAR currAge =
    LASTNONBLANK ( 'Sample'[Age], [Age] )
VAR currYear =
    SELECTEDVALUE ( 'Year Range'[Year] )
VAR lookup =
    LOOKUPVALUE (
        'Sample'[Count],
        'Sample'[Age], currAge,
        'Sample'[Year], currYear
    )
RETURN
    IF (
        lookup <> BLANK (),
        lookup
    )

3. Modify measure to add bin variable to store diff between year and age, use this variable to lookup value from previous records.

Look up 2 = 
VAR currAge =
    LASTNONBLANK ( 'Sample'[Age], [Age] )
VAR currYear =
    SELECTEDVALUE ( 'Year Range'[Year] )
VAR lookup =
    LOOKUPVALUE (
        'Sample'[Count],
        'Sample'[Age], currAge,
        'Sample'[Year], currYear
    )
VAR bin = currYear - currAge
RETURN
    IF (
        lookup <> BLANK (),
        lookup,
        CALCULATE (
            MAX ( 'Sample'[Count] ),
            FILTER ( ALL ( 'Sample' ), [Year] - [Age] = bin )
        )
            + 0
    )

BTW, if you want to summarize all suitable records, please use sum function to replace max function which I used in if statement.

Regards,

Xiaoxin Sheng