Forum Discussion

avcr29's avatar
avcr29
Frequent Visitor
8 years ago
Solved

Quick Measure: Rolling Average per year Help!

Hello!,

 

I need help, i'm using quick measure rolling average and it's calculating correctly but i want to use Year filter, i don't want it to calculate the rolling average taking the average from the prior year, i want it to start the measure for year, so when i calculate it, the result should be like this: 

 

yearmonthx measurex rolling average 
2017Jan71%71%
2017Feb64%68%
2017Mar72%69%
2017Apr72%70%
2017May77%71%
2017Jun75%72%
2017Jul73%72%
2017Aug76%73%
2017Sep68%72%
2017Oct66%72%
2017Nov66%71%
2017Dic66%71%
2018Jan59%59%

 

As you can see in 2018 the rolling average should start calculating without taking in consideration the accumulated of 2017, every year should be like this.

 

The quick measure dax formula is:

x rolling average =

IF(
ISFILTERED('NPS'[Fecha]),
ERROR("Time intelligence quick measures can only be grouped or filtered by the Power BI-provided date hierarchy or primary date column."),
VAR __LAST_DATE = ENDOFMONTH('NPS'[Fecha].[Date])
VAR __DATE_PERIOD =
DATESBETWEEN(
'NPS'[Fecha].[Date],
STARTOFMONTH(DATEADD(__LAST_DATE, -COUNTA(NPS[Fecha]), MONTH)),
__LAST_DATE
)
RETURN
AVERAGEX(
CALCULATETABLE(
SUMMARIZE(
VALUES('NPS'),
'NPS'[Fecha].[Year],
'NPS'[Fecha].[QuarterNo],
'NPS'[Fecha].[Quarter],
'NPS'[Fecha].[MonthNo],
'NPS'[Fecha].[Month]
),
__DATE_PERIOD
),
CALCULATE([x], ALL('NPS'[Fecha].[Day]))
)
)

 

 

i don't know exactly what should i change in the formula or if there's another way.

 

Thanks

  • Hi avcr29

     

    What happens if you try this

     

     

    x rolling average = 
    
    IF(
        ISFILTERED('NPS'[Fecha]),
            ERROR("Time intelligence quick measures can only be grouped or filtered by the Power BI-provided date hierarchy or primary date column."),
    VAR __LAST_DATE = ENDOFMONTH('NPS'[Fecha].[Date])
    VAR __DATE_PERIOD =
        FILTER(
            DATESBETWEEN(
                'NPS'[Fecha].[Date],
                STARTOFMONTH(DATEADD(__LAST_DATE, -COUNTA(NPS[Fecha]), MONTH)),
                __LAST_DATE
                ),YEAR(__LAST_DATE) = YEAR( 'NPS'[Fecha].[Date]))
    RETURN
    AVERAGEX(
    CALCULATETABLE(
    SUMMARIZE(
    VALUES('NPS'),
    'NPS'[Fecha].[Year],
    'NPS'[Fecha].[QuarterNo],
    'NPS'[Fecha].[Quarter],
    'NPS'[Fecha].[MonthNo],
    'NPS'[Fecha].[Month]
    ),
    __DATE_PERIOD
    ),
    CALCULATE([x], ALL('NPS'[Fecha].[Day]))
    )
    )

     

2 Replies

  • Phil_Seamark's avatar
    Phil_Seamark
    Microsoft Employee

    Hi avcr29

     

    What happens if you try this

     

     

    x rolling average = 
    
    IF(
        ISFILTERED('NPS'[Fecha]),
            ERROR("Time intelligence quick measures can only be grouped or filtered by the Power BI-provided date hierarchy or primary date column."),
    VAR __LAST_DATE = ENDOFMONTH('NPS'[Fecha].[Date])
    VAR __DATE_PERIOD =
        FILTER(
            DATESBETWEEN(
                'NPS'[Fecha].[Date],
                STARTOFMONTH(DATEADD(__LAST_DATE, -COUNTA(NPS[Fecha]), MONTH)),
                __LAST_DATE
                ),YEAR(__LAST_DATE) = YEAR( 'NPS'[Fecha].[Date]))
    RETURN
    AVERAGEX(
    CALCULATETABLE(
    SUMMARIZE(
    VALUES('NPS'),
    'NPS'[Fecha].[Year],
    'NPS'[Fecha].[QuarterNo],
    'NPS'[Fecha].[Quarter],
    'NPS'[Fecha].[MonthNo],
    'NPS'[Fecha].[Month]
    ),
    __DATE_PERIOD
    ),
    CALCULATE([x], ALL('NPS'[Fecha].[Day]))
    )
    )