Cannibalization: Set Analysis Filter from Dimensions Values

I have a list of 200 stores, each of them has been openend at different dates, and need to calculate the SUM(SALES) for each of those on the 1st year with this restrictions:

- Sum of sales from customers who have bought something in other stores before the OpenStoreday

I can easily calculate that just by one store, what I don´t know how it to calculate that in 1 measure for all of them.

Let´s say that Store Number 1 Opened on 20171212 (Year, Month, DaY) so STORE_OPENDAY value is 20171212 as well

CANIBALIZATION_VALUE

Sum({<

CLIENT_ID = P({<[INVOICE_DATE]={"<20161212"}>})

*P({<[STORE_NUMBER]={"1"},=>})

>}[SALES])

How should this formula be change in order that INVOICE_DATE could be calculated in a table, depending on the STORE_OPENDAY of each store.

STORE_CODE       OPENDAY          CANIBALIZATION_VALUE

1                               ...                     ...

2                               ...                      ...

.....