search:
Search support.sas.com
KNOWLEDGE BASE
System Requirements
Install Center
Product Documentation
Papers
Samples & SAS Notes
Focus Areas
SUPPORT
License Assistance
Administrator Services
Downloads & Hot Fixes
LEARNING CENTER
Bookstore
Training
Certification
SAS Learning Edition
Higher Education Resources
SAS OnDemand For Academics
COMMUNITY
User Groups
Events
E-Newsletters
Advanced Search
support.sas.com
KNOWLEDGE BASE
SUPPORT
LEARNING CENTER
COMMUNITY
Product Documentation
>
SAS 9.2 Documentation
Print
|
E-mail
|
Bookmark
|
Feedback
SAS(R) High-Performance Forecasting 3.1: User's Guide
PDF
|
Purchase
Contents
About
Acknowledgments
Credits
Documentation
Software
Technical Support
General Information
What’s New in SAS High-Performance Forecasting 3.1
Overview
HPF Procedure Enhancements
HPFDIAGNOSE Procedure Enhancements
HPFENGINE Procedure Enhancements
HPFEVENTS Procedure Enhancements
HPFRECONCILE Procedure Enhancements
Introduction
Overview of SAS High-Performance Forecasting Software
Uses of SAS High-Performance Forecasting Software
Contents of SAS High-Performance Forecasting Software
Experimental Software
About This Book
Chapter Organization
Typographical Conventions
Options Used in Examples
Where to Turn for More Information
Accessing the SAS High-Performance Forecasting Sample Library
Online Help System
SAS Technical Support Services
Related SAS Software
Base SAS Software
SAS/GRAPHSoftware
SAS/STATSoftware
SAS/IMLSoftware
SAS/INSIGHTSoftware
SAS/ORSoftware
SAS/QCSoftware
Other Statistical Tools
References
Procedure Reference
The HPFARIMASPEC Procedure
Overview: HPFARIMASPEC Procedure
Getting Started: HPFARIMASPEC Procedure
Syntax: HPFARIMASPEC Procedure
Functional Summary
PROC HPFARIMASPEC Statement
FORECAST Statement
INPUT Statement
ESTIMATE Statement
Examples: HPFARIMASPEC Procedure
Some HPFARIMASPEC Syntax Illustrations
How to Include ARIMA Models in a Model Selection List
A Generic Seasonal Model Specification Suitable for Different Season Lengths
References
The HPFDIAGNOSE Procedure
Overview: HPFDIAGNOSE Procedure
Getting Started: HPFDIAGNOSE Procedure
Default Settings
The Role of the IDM Statement
Syntax: HPFDIAGNOSE Procedure
Functional Summary
PROC HPFDIAGNOSE Statement
ADJUST Statement
ARIMAX Statement
BY Statement
ESM Statement
EVENT Statement
FORECAST Statement
ID Statement
IDM Statement
INPUT Statement
TRANSFORM Statement
TREND Statement
UCM Statement
Details: HPFDIAGNOSE Procedure
Adjustment Operations
Data Preparation
Functional Transformation
Stationarity Test
ARMA Order Selection
Transfer Functions in an ARIMAX Model
Outliers
Intermittent Demand Model
Exponential Smoothing Model
Unobserved Components Model
Values of Status
Holdout Sample
EVENTS
HPFENGINE
Output Data Sets
ODS Table Names
Examples: HPFDIAGNOSE Procedure
Selection of Input Variables
Selection of Events and Input Variables
Intermittent Demand Series
Exponential Smoothing Model
Unobserved Components Model
References
The HPFENGINE Procedure
Overview: HPFENGINE Procedure
Getting Started: HPFENGINE Procedure
Syntax: HPFENGINE Procedure
Functional Summary
PROC HPFENGINE Statement
ADJUST Statement
BY Statement
CONTROL Statement
EXTERNAL Statement
FORECAST Statement
ID Statement
INPUT Statement
SCORE Statement
STOCHASTIC Statement
Details: HPFENGINE Procedure
Accumulation
Missing Value Interpretation
Adjustment Operations
Diagnostic Tests
Model Selection
Transformations
Parameter Estimation
Missing Value Modeling Issues
Forecasting
Inverse Transformations
Statistics of Fit
Data Set Input/Output
ODS Table Names
ODS Graphics
Examples: HPFENGINE Procedure
The TASK Option
Different Types of Input
Incorporating Events
Using the SCORE Statement
HPFENGINE and HPFDIAGNOSE Procedures
The ADJUST Statement
Multiple Repositories
ODS Graphics
References
The HPFESMSPEC Procedure
Overview: HPFESMSPEC Procedure
Getting Started: HPFESMSPEC Procedure
Syntax: HPFESMSPEC Procedure
Functional Summary
PROC HPFESMSPEC Statement
ESM Statement
Details: HPFESMSPEC Procedure
Smoothing Model Parameter Specification Options
Examples: HPFESMSPEC Procedure
Various Kinds of ESM Model Specifications
Selecting the Best ESM Model
The HPFEVENTS Procedure
Overview: HPFEVENTS Procedure
Getting Started: HPFEVENTS Procedure
Syntax: HPFEVENTS Procedure
Functional Summary
PROC HPFEVENTS Statement
BY Statement
EVENTCOMB Statement
EVENTDATA Statement
EVENTDEF Statement
EVENTDUMMY Statement
EVENTGROUP Statement
EVENTKEY Statement
ID Statement
VAR Statement
Details: HPFEVENTS Procedure
Event Definitions
Using the EVENTKEY Statement
Missing Value Interpretation
Data Set Output
Printed Output
Examples: HPFEVENTS Procedure
Multiple Timing Values in a Single Event versus Using Multiple Events and the EVENTCOMB Statement
Using a DATA Step to Construct an Events Data Set
Preparing a Data Set for PROC HPFENGINE
Using Predefined SAS Event Keywords Directly in Other SAS Procedures
Viewing Dummy Variables by Using the SGPLOT Procedure
References
The HPFEXMSPEC Procedure
Overview: HPFEXMSPEC Procedure
Getting Started: HPFEXMSPEC Procedure
Syntax: HPFEXMSPEC Procedure
Functional Summary
PROC HPFEXMSPEC Statement
EXM Statement
Examples: HPFEXMSPEC Procedure
Various EXM Model Specifications
The HPFIDMSPEC Procedure
Overview: HPFIDMSPEC Procedure
Getting Started: HPFIDMSPEC Procedure
Syntax: HPFIDMSPEC Procedure
Functional Summary
PROC HPFIDMSPEC Statement
IDM Statement
Smoothing Model Suboptions for IDM Statement Options
Details: HPFIDMSPEC Procedure
Smoothing Model Parameter Specification Options
Smoothing Model Forecast Bounds Options
Examples: HPFIDMSPEC Procedure
Various Kinds of IDM Model Specifications
Automatically Choosing the Best Decomposed Demand Model
The HPFRECONCILE Procedure
Overview: HPFRECONCILE Procedure
Getting Started: HPFRECONCILE Procedure
Syntax: HPFRECONCILE Procedure
Functional Summary
PROC HPFRECONCILE Statement
AGGBY Statement
AGGDATA Statement
BY Statement
DISAGGDATA Statement
ID Statement
Details: HPFRECONCILE Procedure
Notation
Top-Down Reconciliation
Bottom-Up Reconciliation
Data Set Input/Output
Examples: HPFRECONCILE Procedure
Reconciling a Hierarchical Tree
Aggregating Forecasts
Disaggregating Forecasts
Imposing Constraints
The HPFSELECT Procedure
Overview: HPFSELECT Procedure
Getting Started: HPFSELECT Procedure
Syntax: HPFSELECT Procedure
Functional Summary
PROC HPFSELECT Statement
DELETE Statement
DIAGNOSE Statement
FORECASTOPTIONS Statement
SELECT Statement
SPECIFICATION Statement
Examples: HPFSELECT Procedure
The INPUTMAP Option
The EVENTMAP Option
The DIAGNOSE Statement
External Models and User-Defined Subroutines
Comparing Forecasts from Multiple External Sources
Input to User-Defined Subroutines
Changing an Existing Selection List
The HPFUCMSPEC Procedure
Overview: HPFUCMSPEC Procedure
Getting Started: HPFUCMSPEC Procedure
Syntax: HPFUCMSPEC Procedure
Functional Summary
PROC HPFUCMSPEC Statement
AUTOREG Statement
BLOCKSEASON Statement
CYCLE Statement
DEPLAG Statement
FORECAST Statement
INPUT Statement
IRREGULAR Statement
LEVEL Statement
SEASON Statement
SLOPE Statement
Examples: HPFUCMSPEC Procedure
Some HPFUCMSPEC Syntax Illustrations
How to Include a UCM in a Model Selection List
How to Create a Generic Seasonal Model Specification
References
The HPF Procedure
Overview: HPF Procedure
Getting Started: HPF Procedure
Syntax: HPF Procedure
Functional Summary
PROC HPF Statement
BY Statement
FORECAST Statement
ID Statement
IDM Statement
Smoothing Model Specification Options for IDM Statement
Details: HPF Procedure
Smoothing Model Parameter Specification Options
Smoothing Model Forecast Bounds Options
Accumulation
Missing Value Interpretation
Diagnostic Tests
Model Selection
Transformations
Parameter Estimation
Missing Value Modeling Issues
Forecasting
Inverse Transformations
Statistics of Fit
Forecast Summation
Comparison to the Time Series Forecasting System
Data Set Output
OUT= Data Set
OUTEST= Data Set
OUTFOR= Data Set
OUTPROCINFO= Data Set
OUTSTAT= Data Set
OUTSUM= Data Set
OUTSEASON= Data Set
OUTTREND= Data Set
Printed Output
ODS Table Names
ODS Graphics
Examples: HPF Procedure
Automatic Forecasting of Time Series Data
Automatic Forecasting of Transactional Data
Specifying the Forecasting Model
Extending the Independent Variables for Multivariate Forecasts
Forecasting Intermittent Time Series Data
Illustration of ODS Graphics
References
Forecasting Details
Forecasting Process Summary
Background
Transactional Data
Time Series Data
Forecasting Models
Forecasts
Forecast Function (Scoring)
Statistics of Fit
Automatic Forecasting Process
Accumulation Step
Interpretation Step
Adjustment Step
Diagnostic Step
Model Selection Step
Parameter Estimation Step
Forecasting Step
Evaluation Step
Performance Step
Forecast Function (Score File) Generation Step
Automatic Forecasting Data
Automatic Forecasting Data Flow
Forecast Scoring Data Flow
Automatic Forecasting Information
Model Specification
Model Selection List
Selected Model Specification
Fitted Model
Forecast Function (Score File)
Automatic Forecasting Information Flow
Forecast Scoring Information Flow
Automatic Forecasting Repositories
Event Repository
Model Specification Repository
Fitted Model Repository
Forecast Results Repository
Score Repository
Automatic Forecasting System Flow
Forecast Scoring System Flow
Automatic Forecasting Archives
References
Forecasting Process Details
Forecasting Process Summary
Parameter Estimation
Model Evaluation
Forecasting
Smoothing Models
Smoothing Model Calculations
Smoothing State and Smoothing Equations
Smoothing State Initialization
Missing Values
Predictions and Prediction Errors
Smoothing Weights
Specifying the Smoothing Weights
Optimizing the Smoothing Weights
Equations for the Smoothing Models
ARIMA Models
ARIMA Model Based Series Decomposition
Unobserved Component Models
Intermittent Models
Intermittent Time Series
Intermittent Series Decomposition and Analysis
Croston’s Method
Average Demand Method
Time-Indexed versus Demand-Indexed Holdout Samples
Automatic Intermittent Demand Model Selection
External Models
External Forecasts
External Forecast Prediction Errors
External Forecast Prediction Bias
External Forecast Prediction Standard Errors
External Forecast Confidence Limits
Series Transformations
Predictions for Transformed Models
Series Diagnostic Tests
Statistics of Fit
References
Using Forecasting Model Score Files and DATA Step Functions
Overview
HPFSCSIG Function
HPFSCSUB Function
Using User-Defined Models
Introduction
Defining and Using a SAS Language Function or Subroutine
Defining and Using a C Language External Function
Input Time Series Keywords
Returned Forecast Component Keywords
Using External Forecasts
Introduction
Specifying and Using an External Model Specification
Index here
Product
Release
SAS High-Performance Forecasting
3.1
SAS/ETS
9.2
SAS Forecast Server
3.1
Type
Usage and Reference
Copyright Date
February 2009
Last Updated
16Jan2009
SAS High-Performance Forecasting 3.1: User's Guide