# Concatenate multiple weeks into one

• ### Question

• In my quest to solve a problem with an average calculation I've made progress, but got stuck again. This probably mainly has to do with my lack of knowledge of MDX, so after searching the net, these forums, and trying numerous options myself, I turn to you guys.

I am trying to calculate a 3 year average on Sales, filtered by stores that had sales over all 3 years, and calculated to a week-weight. So the average week is compared to the average yearly sales, thus returning a week-weight (not sure how to call it really). This avergae needs to be based on weeks 1 to 52 only, so I am trying to filter out all weeks 53.

Part of my query is the following:
```member Measures.Test as
avg(
filter(
{ [Vestiging].[Vestiging Nummer].Members - [Vestiging].[Vestiging Nummer].[All] }
, ([Measures].[IncompletePeriodsYearInd] = 0)
)

,avg(
{
head(ancestor([Datum Verkoop].[Date YWD].CurrentMember, [Datum Verkoop].[Date YWD].[Year]).Lag(1).Children, 52),
head(ancestor([Datum Verkoop].[Date YWD].CurrentMember, [Datum Verkoop].[Date YWD].[Year]).Lag(2).Children, 52),
head(ancestor([Datum Verkoop].[Date YWD].CurrentMember, [Datum Verkoop].[Date YWD].[Year]).Lag(3).Children, 52)
}
, Measures.Omzet

)
)```
This part needs to calculate a yearly total average, but since I get the first 52 weeks, from all 3 years, I get a weekly average. I would like to calculate the average as if it where on the 52 week subset as a whole.

Is such a thing possible? I suspect it should be, but I'm completely lost!