DAX- A measure that count the cumulative percentage (Pareto-Lorenz chart)

[ewelinag17]

Drogi,
potrzebuję pomocy. Chcę utworzyć wykres słupkowy (sumujący wartości w tabeli) i wykres liniowy (miara skumulowanego procenta) Pracuję na zbiorze danych „reklamacje”. Moja miara wygląda tak:

Wykres jest poprawny, jeśli wartości na wykresie słupkowym nie powtarzają się.

I odwrotnie, jeśli wartości są takie same, wykres przedstawia błędną analizę.

Czego brakuje w mojej miarce?

Proszę o pomoc i z góry dziękuję za poświęcony czas.

[dax]

Hi @ewelinag17 , 

It seems that you want to show distinct category in chart, right? If so and if possible could you please inform me more detailed information(such as your logic, expected output and your sample data )? Then I will help you more correctly.

Please do mask sensitive data before uploading.

Thanks for your understanding and support.
Best Regards,
Zoe Zhi

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[ewelinag17]

@dax 

An example data table is shown below.

I create a graph from the data.

I create a line chart with measure:

pareto-l =

VAR TotalIndeks = CALCULATE(SUM('Table'[Indeks.2]),ALLSELECTED('Table'))

VAR CurrentIndeks = SUM('Table'[Indeks.2])

VAR SummarizedTable =

SUMMARIZE(

    ALLSELECTED('Table'),

    'Table'[pojemnosc],

    "Indeks.2",SUM('Table'[Indeks.2])

)

VAR CumulativeSum =

SUMX(

    FILTER(SummarizedTable,[Indeks.2] >= CurrentIndeks),

    [Indeks.2]

)

RETURN

DIVIDE(CumulativeSum,TotalIndeks)

The chart is good, but I need to calculate the cumulative percentage. 

190 - 33,3%

130 - 61.9%

125 - 76.2%

300 - 85,7%

180 - 90,4%

250 - 95,2%

90 - 100%

I guess a random variable is missing(rand??), but I'm new to Power BI and don't know how to do this.

Thank you in advance for your help.

[dax · Accepted solution]

Hi @ewelinag17 ,

You could try M code to create a new column

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMrQ0UNJRMlSK1QGyjZHYyOIWCLaxAZK4kSl29TjMRNFrTAQbyRwkppEpDquQnWNMwJmxAA==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type text) meta [Serialized.Text = true]) in type table [pojemnosc = _t, Indeks.2 = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"pojemnosc", Int64.Type}, {"Indeks.2", Int64.Type}}),
    #"Grouped Rows" = Table.Group(#"Changed Type", {"pojemnosc"}, {{"Count", each Table.RowCount(_), type number}, {"all", each _, type table [pojemnosc=number, Indeks.2=number]}}),
    #"Sorted Rows" = Table.Sort(#"Grouped Rows",{{"Count", Order.Ascending}}),
    #"Added Index" = Table.AddIndexColumn(#"Sorted Rows", "Index", 0, 1),
    #"Expanded all" = Table.ExpandTableColumn(#"Added Index", "all", {"pojemnosc", "Indeks.2"}, {"pojemnosc.1", "Indeks.2"}),
    #"Removed Columns" = Table.RemoveColumns(#"Expanded all",{"Count", "pojemnosc.1"})
in
    #"Removed Columns"

 You could refer to my sample for details.

Best Regards,
Zoe Zhi

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[ewelinag17]

@dax 

Great! But if I add a column and filter relative to the new column, the chart won't be stacked. 

That's enough for now, but in the future this file will grow by a few columns. And then how to filter on new columns?

[dax]

Hi @ewelinag17 , 

As I know, the slicer will affect on measure, it seem not affect on calculated column, so you need to use measure to achieve this goal.

Best Regards,
Zoe Zhi

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.