FILTER and PREVIOUSMONTH function DAX , how to combine

[Eric-De]

Under powerautomate I used the PowerBI dataset to export some values under excell file, and I need to filter on the previousmonth only. 

For this I have create a calendar table to compare with the date closed under the table Tickets but when I implent the 

FILTER ( Tickets, Tickets[closed] = PREVIOUSMONTH(Calendar[Date])) 

instead of

FILTER ( Tickets, Tickets[closed] < DATE(YEAR(today()),MONTH(today()),1) && Tickets[closed] >= DATE(YEAR(today()),MONTH(today())-1,1) )

that return nothing

Here my script DAX who running

Can you help me to use corectly the PREVIOUSMONTH function for this ? 

// DAX Query

DEFINE

  VAR vType =   TREATAS({"CPU", "MEM","DISK"}, Tickets[Service])

EVALUATE

  SUMMARIZECOLUMNS

  (

    ROLLUPADDISSUBTOTAL

    (

      Tickets[Service], "Type", vType,

      TREATAS ( { "OK" }, Tickets[Status] ),

  FILTER( Tickets , RELATED(ServiceOffering[Service Offering]) ="TOP" ),

  FILTER ( Tickets, Tickets[closed] < DATE(YEAR(today()),MONTH(today()),1) && Tickets[closed] >= DATE(YEAR(today()),MONTH(today())-1,1) )

    ),

    "Total Qty", COUNT(Tickets[Target])

  )

ORDER BY

  [Type] DESC, [Service]

[Eric-De]

Hello 

So need to create a table date for this 

CALENDAR (

    DATE ( YEAR ( MIN ( Tickets[Closed] ) ), 1, 1 ),

    DATE ( YEAR ( MAX ( Tickets[Closed] ) ), 12, 31 )

)

and because column containt date and time problably need o separate the date and time to group result by month

But in all case thank for your reply 

[lbendlin · Accepted solution]

DEFINE
  VAR vType =   TREATAS({"CPU", "MEM","DISK"}, Tickets[Service])
 
EVALUATE
  SUMMARIZECOLUMNS
  (
    ROLLUPADDISSUBTOTAL
    (
      Tickets[Service], "Type", vType,
      TREATAS ( { "OK" }, Tickets[Status] ),
  FILTER( Tickets , RELATED(ServiceOffering[Service Offering]) ="TOP" ),
  FILTER ( Dates, PREVIOUSMONTH(Dates[Date] )
    ),
    "Total Qty", COUNT(Tickets[Target])
  )

This assumes you have a Dates table in your data model and that it is joined to the Close Date.