Calculated Sum of values filtered by date and selected value

[Anonymous]

Is there anyway to achieve this?

I want to measure some values from a Table
I want to see the number of events from yesterday, last week and last month compared with today's events, current week and current month to see if we had an increase or decrease in the number of events

I have already achieve that with this fomula 

Yesterday = CALCULATE(SUM(Alarmas_Estacion[Eventos]),FILTER(Alarmas_Estacion,Alarmas_Estacion[Fecha sin Hora] = TODAY()-1)) (I have a similar for last week and month)

But this sums all the values in the table and I want to see this values filtered by the "Estacion" name 

I want to see how many events does the "estacion" DJM305 had yesterday vs todays 
and if on this table the users selects "DJM369" the values from events change to the "estacion" selected 

similar to PowerApps ThisIte.SelectedText.Value 

[v-kelly-msft · Accepted solution]

Hi  @Anonymous ,

Create a measure as below:

Yesterday =
CALCULATE (
    SUM ( Alarmas_Estacion[Eventos] ),
        FILTER ( ALL(Alarmas_Estacion), Alarmas_Estacion[Fecha sin Hora] = TODAY () - 1&&Alarmas_Estacion[Estacion]="DJM305" )
    )
)

Best Regards,
Kelly

Did I answer your question? Mark my post as a solution!

[jdbuchanan71]

@Anonymous 

Try adding KEEPFITLERS to your measure.

Yesterday =
CALCULATE (
    SUM ( Alarmas_Estacion[Eventos] ),
    KEEPFILTERS (
        FILTER ( Alarmas_Estacion, Alarmas_Estacion[Fecha sin Hora] = TODAY () - 1 )
    )
)

[Anonymous]

It works but I can only used in one DAX formula

If I use KEEPFILTERS in Yesta¿erday formula then I cannot used it in Todays or Last Month 
it creates a circular reference between them

[jdbuchanan71]

@Anonymous 

I beleive that is becaue you are using the FILTER in your meaure.  Take a look at this article about working with weeks in time intelligence funcitons.

https://www.daxpatterns.com/week-related-calculations/

It is part of a larger series on time patterns that you might find useful.

https://www.daxpatterns.com/time-patterns/