RANKX Expression for 3 Columns

[Matt1979]

Hi everyone, first post.  Using the sample data below (call the table RANKX Test), I want to write a RANKX measure that ranks the Attribute column by the Value column, for each ID column value:

ID	Attribute	Value
111	A	60
111	B	50
112	A	50
112	B	70
113	A	20
113	B	40

I know how to rank attribute by value, and create a visual level filter for the top ranked attribute:

Rank Attributes = IF(RANKX(ALLSELECTED('RANKX Test'[Attribute]),CALCULATE(SUM('RANKX Test'[Value])),,ASC,Dense)

                                 <=1,1,0)

A table with the Attribute column and a column with the COUNT of Attribute values looks like this:

Attribute         Count of Attribute

A                     3

I need to add the ID column to the measure, so that the attributes are ranked for each ID.  I would want the table above to look like this since A would be the top ranked attribute for two IDs and B would be the top ranked attribute for one ID:

Attribute             Count of Attribute

A                         2

B                         1

Appreciate any help on this!

[Jihwan_Kim · Accepted solution]

Hi,

I am not sure how your datamodel looks like, but please check the below picture and the attached pbix file.

Count of attribute expected result: =
VAR _currentrowattribute =
    MAX ( Data[Attribute] )
RETURN
    IF (
        HASONEVALUE ( Data[Attribute] ),
        COUNTROWS (
            FILTER (
                ADDCOLUMNS (
                    ALL ( Data ),
                    "@Rank_partion_ID",
                        RANK (
                            skip,
                            ALL ( Data ),
                            ORDERBY ( Data[Value], ASC ),
                            PARTITIONBY ( Data[ID] )
                        )
                ),
                Data[Attribute] = _currentrowattribute
                    && [@Rank_partion_ID] = 1
            )
        )
    )

[Matt1979]

Thanks!  That works...