The goal of the data warehouse is to provide analysis of data via aggregates over dimensions using OLAP cubes.
SQL Server 2005 and SQL Server 2008 do this job excellently.
Getting a historic snapshot day by day is difficult.
Some thoughts: you can introduce a day dimension and use it for filtering for a particular day or prior. Naturally this would add overhead. You would have to tag each fact with this dimension.
You can also take a snapshot of the data warehouse database daily and run OLAP cube processing on them when needed like end of months or end of quarters.
Let us know how you decide.
Kalman Toth - Database, Data Warehouse & Business Intelligence Architect
BI Workshop NYC SSAS, SSIS, SSRS - April 21-24: http://www.sqlusa.com/