Filter Top N% Column Values by Other Column Values

Anonymous file attached for your reference, enjoy!!

Anonymous I think this will do it, I guess this is what you are looking for:

Anonymous  sure no problem. Can you please provide some data and desired output please.

Anonymous  

I don't think PBI has anything which utilizes % as TOPN

What you want can be returned through a DAX table function but can this be done with a measure. looping AlexisOlson  and parry2k  attached a pbix for your refrerence

DAX Table function that returns the derived table, but offcourse that is not desired. The foloowing DAX query returns if a the same PIVOT table was filtered in excel

Measure =
VAR _1 =
    ADDCOLUMNS ( 'Table 1', "_pct", [pct] )
VAR _2 =
    ADDCOLUMNS (
        _1,
        "rank",
            RANKX (
                FILTER (
                    _1,
                    'Table 1'[Local Product Group Unit]
                        = EARLIER ( 'Table 1'[Local Product Group Unit] )
                ),
                [_pct],
                ,
                DESC
            )
    )
VAR _3 =
    ADDCOLUMNS (
        _2,
        "runningPct",
            SUMX (
                FILTER (
                    _2,
                    'Table 1'[Local Product Group Unit]
                        = EARLIER ( 'Table 1'[Local Product Group Unit] )
                        && [rank] <= EARLIER ( [rank] )
                ),
                [_pct]
            )
    )
VAR _4 =
    ADDCOLUMNS (
        ADDCOLUMNS ( _3, "is60", IF ( [runningPct] >= 0.6, 0, [runningPct] ) ),
        "is60_2", IF ( [runningPct] >= 0.6, [runningPct], 0 )
    )
VAR _5 =
    ADDCOLUMNS (
        _4,
        "rank2",
            IF (
                [is60] = 0,
                RANKX (
                    FILTER (
                        _4,
                        'Table 1'[Local Product Group Unit]
                            = EARLIER ( 'Table 1'[Local Product Group Unit] )
                    ),
                    [is60_2],
                    ,
                    asc,
                    DENSE
                )
            )
    )
VAR _6 =
    FILTER ( _5, [is60] <> 0 )
VAR _7 =
    FILTER ( _5, [rank2] = 2 )
VAR _8 =
    NATURALINNERJOIN (
        'Table 1',
        SUMMARIZE ( UNION ( _6, _7 ), [Local Product Group Unit], [Supplier] )
    )
RETURN
    _8

Anonymous not fully clear what you are looking for. can you provide an example of the output?

Anonymous Given the data, is this the expected output