SAS/ETS User's Guide

Getting Started with Time Series Forecasting

This chapter outlines the forecasting process and introduces the major windows of the system through three example sessions.

The first example, beginning with the section "The Time Series Forecasting Window," shows how to use the system for fully automated forecasting of a set of time series. This example also introduces the system's features for viewing data and forecasts through tables and interactive graphs, and shows how to save and restore forecasting work in SAS catalogs.

The second example, beginning with the section "Develop Models Window," introduces the features for developing the best forecasting models for individual time series. The chapter concludes with an example showing how to create dating variables for your data in the form expected by the system.

After working through the examples in this chapter, you should be able to

select a data set of time series to work with and specify its periodicity and time ID variable
use the automatic forecasting model selection feature to create forecasting models for the variables in a data set
produce and save forecasts of variables in a data set
examine your data and forecasts as tables of values and through interactive graphs
save and restore your forecasting models using project files in a SAS catalog and edit project information
use some of the model development features to fit and select forecasting models for individual time series variables.

This chapter introduces these topics and will help you get started using the system. Later chapters present these topics in greater detail and document more advanced features and options.

Getting Started with Time Series Forecasting

The Time Series Forecasting Window

Outline of the Forecasting Process

The Input Data Set

Automatic Model Fitting Window

Produce Forecasts Window

The Forecast Data Set

Forecasting Projects

Develop Models Window

Model Viewer