As Grant said, this is different in Azure. You treat the system as a database service, meaning you connect to a database and see the objects, but that's it. There isn't any control over, nor need to worry about, the way in which the physical structure of the database is organized.

I would guess your application not doesn't really need to know indexes are separate from data in filegroups. Someone made that decision, and potentially a setup process/program does this, but you would ignore that in Azure SQL Database. The database gets created and then you just put objects and data inside it.

It works like this: http://www.sqlservercentral.com/articles/Azure+SQL+Platform+as+Service+%28PaaS%29/128108/