SAS® High-Performance Forecasting
Large-scale forecasting is a challenge for many industries. Retailers forecast weekly sales for thousands of items at hundreds, or thousands, of stores. Utility companies forecast electricity demand at thousands of points in their customer network, in few-minute intervals. Generating customized models for millions of forecasts would require an army of skilled statisticians.
- Generate forecasts faster for high-value and time-sensitive decision making.
- Forecast at more granular levels of detail, in finer time increments.
- Take advantage of a highly scalable and reliable analytics infrastructure.
- Efficiently process large-scale, hierarchically structured, time-stamped data sets.
- Process multiple hierarchy levels in a single pass of the data set.
- Automatically generate time series from the time-stamped data with forecasts produced for them in one step.
- For typical time series, the best-performing forecast is chosen automatically from the following smoothing models: Simple, linear, damped trend, seasonal (additive and multiplicative), Winters' method (additive and multiplicative).
- Transformed versions of the models include log, square root, logistic and Box-Cox.
- Forecast time series data, whose observations are equally spaced by a specific time interval (for example, monthly, weekly).
- Forecast transactional data, whose observations are not spaced with respect to any particular time interval.