CESM Procedure

The CESM procedure generates forecasts by using exponential smoothing models with optimized smoothing weights for one or more time series. Time series data have observations that are equally spaced by a specific time interval (for example, monthly or weekly).

PROC CESM can also generate these forecasts for transactional data, whose observations are not spaced with respect to any particular time interval. The procedure accumulates transactional data on the basis of a specified (or default) time interval to form a time series prior to modeling and forecasting. Typical examples of transactional data are internet, inventory, sales, and other similar data.

For typical time series, you can use the following smoothing models: simple, double, linear, damped trend, seasonal, and Winters method (additive or multiplicative). In addition, transformed versions of the following models are provided: log, square root, logistic, and Box-Cox.

The CESM procedure writes the following to output data tables: the time series that are extrapolated by the forecasts, the series summary statistics, the forecasts and confidence limits, and the parameter estimates.

PROC CESM requires SAS Cloud Analytic Services (CAS) in order to run, and it does the following: