wrong averaged weight

[Anonymous]

I am trying to get a weighted average (as a total), that is, for each customer, :

sum(value * size)     /  sum(size)

Tthe value is just a measure that is (SumPerCustomer / Max_PerCustomer)

...If a customer has (3 , 1,7 ) the measure shall be 11 / 7

I get the desired result for the value but not for the averaged weight:

VAR Max_PerCustomer =

MAXX(

    FILTER(

    SUMMARIZE(

        Revenue

        ,'Customer'[Customer]

        ,'Center'[Center]

    )

    , SUM(Revenue[Revenue]) > 0

)

    , SUM(Revenue[Revenue])

)

VAR SumPerCustomer =

CALCULATE(

    SUM(Revenue[Revenue])

    , FILTER(

    SUMMARIZE(

        Revenue

        ,'Customer'[Customer]

        ,'Center'[Center]

    )

    , SUM(Revenue[Revenue]) > 0

)

)

RETURN

SWITCH(

    TRUE()

    ,ISINSCOPE('Customer’[Customer]) , DIVIDE(SumPerCustomer, Max_PerCustomer)

    ,DIVIDE ( SUMX('Customer’,    DIVIDE(SumPerCustomer, Max_PerCustomer) * SumPerCustomer),  SUMX('Customer’, SumPerCustomer),0)

)

Any help is appreciated, I can't figure why it does the wrong calculation

[ryan_mayu]

@Anonymous 

could you pls provide the sample data and expected output?

[amitchandak]

@Anonymous , try to add values and check

measure =
var _sel = selectedvalue(Param[Param])
var _cnt =distinctcount(Geography[locations])
var _1= CALCULATE([Sales],TOPN(_sel,all(Geography[locations]),[Sales],DESC),VALUES(Geography[locations]))
var _2= CALCULATE([Sales],TOPN(_cnt,all(Geography[locations]),[Sales],asc),VALUES(Geography[locations]))
return
if(_sel =0 || not(isfiltered(Param[Param])) , _2, _1)

calculate(
SWITCH(

TRUE()

,ISINSCOPE('Customer'[Customer]) , DIVIDE(SumPerCustomer, Max_PerCustomer)

,DIVIDE ( SUMX('Customer', DIVIDE(SumPerCustomer, Max_PerCustomer) * SumPerCustomer), SUMX('Customer', SumPerCustomer),0)

), values('Customer'[Customer]))

[Anonymous]

I can' quiet follow , what do you mean?

[amitchandak]

@Anonymous , values is group data at a level

Can you share sample data and sample output in table format? Or a sample pbix after removing sensitive data.