SAS 2012

Application of Time Series Clustering Using SAS® Enterprise Miner™ for a Retail Chain
Much of the data that are generated in the operational side of a business have a built-in time dimension. One of the challenges of doing data mining using such time-series data is the complexity of handling a large number of time series. Time series clustering provides a way to reduce the complexity by categorizing large number of time series into a smaller subset such that series within each subset are relatively homogenous but series between subsets are heterogeneous.SAS® has recently introduced new nodes for finding similarities between the time series and to forecast their future trajectories. In this paper we demonstrate clustering of store-level revenue over time and how profiling of such clusters generate additional business insights.

Walt Disney World Dolphin Resort