Forum Discussion

S3's avatar
S3
Helper III
3 years ago
Solved

PreviousDay for First Date in Dataset

Hello,  I have a dataset which begins on the 01st of January 2022 and ends on the 31st of July 2022.  I also have a date slicer the user can choose from and a graph showing the value of previo...
Need Help
  • v-yalanwu-msft's avatar
    v-yalanwu-msft
    3 years ago

    Hi, S3 ;

    Try it.

    Previous Period Value = 
var _a= CALCULATE([Value Measure]
	                             ,DATESBETWEEN(
									 Dates[Date],
									 [Start of Previous Period],
				    				 [End of Previous Period]),
									 ALL(Dates))
var _diff2=DATEDIFF( CALCULATE(MIN('Dates'[Date]),ALLSELECTED(Dates)),CALCULATE(MAX('Dates'[Date]),ALLSELECTED(Dates)),DAY)
var _diff1=DATEDIFF( CALCULATE(MIN('Dates'[Date]),ALL(Dates)),CALCULATE(MIN('Dates'[Date]),ALLSELECTED(Dates)),DAY)
return IF(CALCULATE(SUM('Table'[value]),ALL('Table'))=_a||_diff1<_diff2,BLANK(), _a)

    The final show:


    Best Regards,
    Community Support Team _ Yalan Wu
    If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

S3's avatar
S3
Helper III
3 years ago

Hello lbendlin, 

You've once helped me with a similar concept regarding previous month value. I've tried changing in your previous measure for this case, however I simply got the value of the same date and not the previous. 
This case also has an aditional point of having a selected period of time, and wishing to get N/A when the number of current period's days is greater than the number of the previous days'.

Do you have an idea about that please? 

  • lbendlin's avatar
    lbendlin
    Super User
    3 years ago

    Keep in mind that if you want to use time intelligence functions then your calendar table must be contiguous and must also cover the result of the calculation.