About this paper
With any forecasting challenge, the most important first step is to understand the structure of your time series data. But when dealing with a large number of time series, it is impractical to manually review, diagnose and construct a customized forecasting model for each series. This is where the automatic, large-scale forecasting capabilities of SAS Forecast Server can help.
In the ideal forecasting world, good forecasting results are easily achieved through an automated forecasting engine and one forecast modeling strategy. Unfortunately, given the range of time series patterns, one modeling strategy for all the different types of time series will NOT produce the most accurate forecast.
So how can you manage all the different patterns of your time series data? The answer is time series segmentation, which divides time series data into distinct types or segments based on the underlying properties. It is one of the most important first steps in the forecasting process because it allows you to apply a customized forecast modeling strategy to each time series segment.
This paper looks at two real-world examples where SAS customers are using time series segmentation to improve their forecasting processes. It also introduces the SAS Forecast Server Client web interface and shows how easily you can make time series segmentation an integrated part of your forecasting process.
SAS is the leader in analytics. Through innovative analytics, business intelligence and data management software and services, SAS helps customers at more than 83,000 sites make better decisions faster. Since 1976, SAS has been giving customers around the world THE POWER TO KNOW®.