different DAX output between SSAS vs Import

[Kagechiyo]

hello,

i would like to make a accumlative / pareto chart by year-month, in clustered chart.

i also make 2 reports :

A - first i make using import data from DB source directly

then

B - i make the report using SSAS from the DB source

so i make the DAX as follow :

//--------

accumulate unit=

var currentmon =

SELECTCOLUMNS('VW_FactDetailUnit',"currmonth",'VW_FactDetailUnit'[month])

var sumtabledax =

SUMMARIZE(

ALLSELECTED('VW_FactDetailUnit'),

'VW_PC_FactDetailVIN'[dDate],

"kumakun", SUM('VW_FactDetailUnit'[Qty]))

var cummulativesum =

SUMX(

FILTER(sumtabledax, MONTH('VW_FactDetailUnit'[dDate]) <= currentmon),

[kumakun]

)

return

cummulativesum

//-------------

but the result is not the same..

A - the result is OK it can shown correctly.

but

B - is showing error msg as follow :

"The expression refers to multiple columns. Multiple column cannot be converted to a scalar value."

is there any step(s) that i missed for the B report?

thank you in advance for your kind attention / time.

best regards,

[Anonymous]

Regarding the use of SUMMARIZE... I encourage you to read this: All the secrets of SUMMARIZE - SQLBI, and then regardnig the use of ALLSELECTED, I encourage you to read this: The definitive guide to ALLSELECTED - SQLBI

By the way, your "currentmon" can't return more than 1 value because you then use it in an expression with "<=" and this requires a scalar. I don't think SELECTCOLUMNS is the right function to use in this context. There are much better ones. For example, SELECTEDVALUE (supported in DirectQuery).

[amitchandak]

@Kagechiyo , Try a measure like

var currentmon =
maxx(allselected('VW_FactDetailUnit'),'VW_FactDetailUnit'[month])

var sumtabledax =
SUMMARIZE(
ALLSELECTED('VW_FactDetailUnit'),
'VW_PC_FactDetailVIN'[dDate],
"kumakun", SUM('VW_FactDetailUnit'[Qty]))

var cummulativesum =
SUMX(
FILTER(sumtabledax, MONTH('VW_FactDetailUnit'[dDate]) <= currentmon),
[kumakun]
)

return
cummulativesum