Calculate % increase from last Period (Month)

cristianml 

Wow, no kidding that your data model is complex. However, having a dim_date table is cruicial for any data model, because the dim_date table should be the main source to slice and calculate around your fact table, in this case, it is your period table.

Also, creating a dim_date table won't add complexity to your data model at all, because it will always be a one(dim_date) to many (other fact tables) relationship, and there is no ambiguity.

After you have the dim_date table, just write:

calculate(actual, previsoumonth(dim_date))

that should work.

Hope this helps

Those're actually two separate measures.

Measure 1: Last Months Actual LCR

Actual LCR LM =
VAR last_date = LASTDATE(ALL('Resource Actual'[Date Column]))
VAR current_date = LASTNONBLANK(DateDimension[Date],[Actual LCR])
VAR ly_date = NEXTDAY(PREVIOUSMONTH(current_date))
VAR date_context =
    DATESBETWEEN(
        DateDimension[Date]
        ,NEXTDAY(PREVIOUSMONTH(current_date))
        ,LASTDATE(current_date)
    )
VAR result = CALCULATE([Actual LCR],DATEADD(date_context,-1,MONTH))
RETURN
result

Measure 2: Variance

Variance = DIVIDE(
   [Actual LCR] - [Actual LCR LM]
   ,[Actual LCR LM]
) * SIGN([Actual LCR LM])