So your approach is a good one: you are using Analytic Views to get great performance from large star schemas. A Window Function will definitely not work here because Window Functions unfortunately execute in the row store this will transfer a truck load of data and be very slow.
Normally in this case I recommend building a calc view on top of both Analytic Views and then using Union with Constant Values.
Is there any reason why you can't do this rather than a join on two subqueries?
If you really need a join then I recommend you create a scripted calc view and use CE Functions. I've used this in your scenario and it performs well.
Hope this helps.
John