Using SAS High-Performance Forecasting Software

Duration: 2.0 days

Audience

This course is designed for professional forecasters and business analysts who need to produce forecasts for large numbers of time series.

Course Description

This course enables you to make accurate forecasts quickly and automatically, giving you the power to confidently plan your business operations. This course focuses on the following areas: creating programs that provide forecasts as output files or SAS data sets and structuring the programs and output so that they can be incorporated into a corporate forecasting system; assessing forecast performance and making decisions about the adequacy of initial forecasts, as well as the adequacy of forecasts that are being dynamically produced; and manually overseeing forecasting or devising automated strategies to determine when a forecast model should be updated.

Prerequisites

Before attending this course, you should

have experience with a Windows computing environment
have experience using a product such as Microsoft Excel to enter or transfer data and to perform elementary analyses such as computing row and column totals and averages and producing charts and plots
have taken SAS Programming I: Essentials or have equivalent experience.

The ideal student should have a bachelor's degree or equivalent experience in business, computer science, or a subject matter field that includes quantitative analysis. Forecasting experience is not required.

Course Contents

Introduction to Time Series Forecasting

motivation: the large-scale forecasting problem
time series characteristics
time series components

Exploring and Processing Time-Stamped Data

identifying systematic variation in the data
accumulation: transforming transactional data into time series data
aggregation: building the data hierarchy
two feasible data layouts for SAS Forecast Server
a SAS toolbox for exploring time series data

The Design of SAS High-Performance Software

the design of SAS High-Performance Forecasting software
functionality overview: system inputs and automatic model selection

Forecast Models

the exponential smoothing family of models
the ARIMA family of models
other families of models supported: IDM, UCM
using models in SAS Forecast Server

Diagnosing and Selecting Models Automatically

model selection lists and automatic model selection
the HPFDIAGNOSE procedure: automatic model specification

Creating Custom Models and Managing Model Lists

creating custom models: the HPFxxxSPEC procedures
creating and managing model selection lists: the HPFSELECT procedure

Using the Events Functionality

creating events: the HPFEVENTS procedure

Producing, Assessing, and Modifying Forecasts

preparing to generate forecasts: accumulation and aggregation of data
preparing to generate foreccasts: creating custom models, diagnosing models, and building a model selection list
generating forecasts: effects with the HPFRECONCILE procedure and assessing forecast accuracy
rolling the system forward in time: accommodating new data
assessing changes in forecast accuracy and modifying forecasts

The %HPFRECON Macro

generating reconciled forecasts using the %HPFRECON macro

A Useful Application of the INTNX Function

using the INTNX function to extend ID values

Software Addressed

This course addresses the following software product(s): SAS Forecast Server, SAS/ETS, SAS/STAT.

Share Your Thoughts

Are there additional topics you'd like for this course to address? Would you like for this course to be offered at another training facility? Let us know by adding to our Interest List.

This page was created using SAS software.