You are previewing Applied Data Mining for Forecasting Using SAS(R).
O'Reilly logo
Applied Data Mining for Forecasting Using SAS(R)

Book Description

Written for forecasting practitioners, engineers, statisticians, and economists, the book details how to select useful candidate input variables for time series regression models in environments when the number of candidates is large and identifies the correlation structure between selected candidate inputs and the forecast variable.

Table of Contents

  1. Cover Page
  2. Title Page
  3. Copyright Page
  4. Contents
  5. Preface
  6. Chapter 1 - Why Industry Needs Data Mining for Forecasting
    1. 1.1 - Overview
    2. 1.2 - Forecasting Capabilities as a Competitive Advantage
    3. 1.3 - The Explosion of Available Time Series Data
    4. 1.4 - Some Background on Forecasting
    5. 1.5 - The Limitations of Classical Univariate Forecasting
    6. 1.6 - What is a Time Series Database?
    7. 1.7 - What is Data Mining for Forecasting?
    8. 1.8 - Advantages of Integrating Data Mining and Forecasting
    9. 1.9 - Remaining Chapters
  7. Chapter 2 - Data Mining for Forecasting Work Process
    1. 2.1 - Introduction
    2. 2.2 - Work Process Description
      1. 2.2.1 - Generic Flowchart
      2. 2.2.2 - Key Steps
    3. 2.3 - Work Process with SAS Tools
      1. 2.3.1 - Data Preparation Steps with SAS Tools
      2. 2.3.2 - Variable Reduction and Selection Steps with SAS Tools
      3. 2.3.3 - Forecasting Steps with SAS Tools
      4. 2.3.4 - Model Deployment Steps with SAS Tools
      5. 2.3.5 - Model Maintenance Steps with SAS Tools
      6. 2.3.6 - Guidance for SAS Tool Selection Related to Data Mining in Forecasting
    4. 2.4 - Work Process Integration in Six Sigma
      1. 2.4.1 - Six Sigma in Industry
      2. 2.4.2 - The DMAIC Process
      3. 2.4.3 - Integration with the DMAIC Process
    5. Appendix: Project Charter
  8. Chapter 3 - Data Mining for Forecasting Infrastructure
    1. 3.1 - Introduction
    2. 3.2 - Hardware Infrastructure
      1. 3.2.1 - Personal Computers Network Infrastructure
      2. 3.2.2 - Client/Server Infrastructure
      3. 3.2.3 - Cloud Computing Infrastructure
    3. 3.3 - Software Infrastructure
      1. 3.3.1 - Data Collection Software
      2. 3.3.2 - Data Preparation Software
      3. 3.3.3 - Data Mining Software
      4. 3.3.4 - Forecasting Software
      5. 3.3.5 - Software Selection Criteria
    4. 3.4 - Data Infrastructure
      1. 3.4.1 - Internal Data Infrastructure
      2. 3.4.2 - External Data Infrastructure
    5. 3.5 - Organizational Infrastructure
      1. 3.5.1 - Developers Infrastructure
      2. 3.5.2 - Users Infrastructure
      3. 3.5.3 - Work Process Implementation
      4. 3.5.4 - Integration with IT
  9. Chapter 4 - Issues with Data Mining for Forecasting Application
    1. 4.1 - Introduction
    2. 4.2 - Technical Issues
      1. 4.2.1 - Data Quality Issues
      2. 4.2.2 - Data Mining Methods Limitations
      3. 4.2.3 - Forecasting Methods Limitations
    3. 4.3 - Nontechnical Issues
      1. 4.3.1 - Managing Forecasting Expectations
      2. 4.3.2 - Handling Politics of Forecasting
      3. 4.3.3 - Avoiding Bad Practices
      4. 4.3.4 - Forecasting Aphorisms
    4. 4.4 - Checklist “Are we Ready?”
  10. Chapter 5 - Data Collection
    1. 5.1 - Introduction
    2. 5.2 - System Structure and Data Identification
      1. 5.2.1 - Mind-Mapping
      2. 5.2.2 - System Structure Knowledge Acquisition
      3. 5.2.3 - Data Structure Identification
    3. 5.3 - Data Definition
      1. 5.3.1 - Data Sources
      2. 5.3.2 - Metadata
    4. 5.4 - Data Extraction
      1. 5.4.1 - Internal Data Extraction
      2. 5.4.2 - External Data Extraction
    5. 5.5 - Data Alignment
      1. 5.5.1 - Data Alignment to a Business Structure
      2. 5.5.2 - Data Alignment to Time
    6. 5.6 - Data Collection Automation for Model Deployment
      1. 5.6.1 - Differences Between Data Collection for Model Development and Deployment
      2. 5.6.2 - Data Collection Automation for Model Deployment
  11. Chapter 6 - Data Preparation
    1. 6.1 - Overview
    2. 6.2 - Transactional Data Versus Time Series Data
    3. 6.3 - Matching Frequencies
      1. 6.3.1 - Contracting
      2. 6.3.2 - Expanding
    4. 6.4 - Merging
    5. 6.5 - Imputation
    6. 6.6 - Outliers
    7. 6.7 - Transformations
    8. 6.8 - Summary
  12. Chapter 7 - A Practitioner's Guide of DMM Methods for Forecasting
    1. 7.1 - Overview
    2. 7.2 - Methods for Variable Reduction
      1. Traditional Data Mining
      2. Time Series Approach
    3. 7.3 - Methods for Variable Selection
      1. Traditional Data Mining
      2. Example for Variable Selection
      3. Variable Selection Based on Pearson Product-Moment Correlation Coefficient
      4. Variable Selection Based on Stepwise Regression
      5. Variable Selection Based on the SAS Enterprise Miner Variable Selection Node
      6. Variable Selection Based on the SAS Enterprise Miner Partial Least Squares Node
      7. Variable Selection Based on Decision Trees
      8. Variable Selection Based on Genetic Programming
      9. Comparison of Data Mining Variable Selection Results
    4. 7.4 - Time Series Approach
    5. 7.5 - Summary
  13. Chapter 8 - Model Building: ARMA Models
    1. Introduction
    2. 8.1 - ARMA Models
      1. 8.1.1 - AR Models: Concepts and Application
      2. 8.1.2 - Moving Average Models: Concepts and Application
      3. 8.1.3 - Auto Regressive Moving Average (ARMA) Models
      4. Appendix 1: Useful Technical Details
      5. Appendix 2: The “I” in ARIMA
  14. Chapter 9 - Model Building: ARIMAX or Dynamic Regression Modes
    1. Introduction
    2. 9.1 - ARIMAX Concepts
    3. 9.2 - ARIMAX Applications
    4. Appendix: Prewhitening and Other Topics Associated with Interval-Valued Input Variables
  15. Chapter 10 - Model Building: Further Modeling Topics
    1. Introduction
    2. 10.1 - Creating Time Series Data and Data Hierarchies Using Accumulation and Aggregation Methods
      1. Introduction
      2. Creating Time Series Data Using Accumulation Methods
      3. Creating Data Hierarchies Using Aggregation Methods
    3. 10.2 - Statistical Forecast Reconciliation
    4. 10.3 - Intermittent Demand
    5. 10.4 - High-Frequency Data and Mixed-Frequency Forecasting
      1. High-Frequency Data
      2. Mixed-Interval Forecasting
    6. 10.5 - Holdout Samples and Forecast Model Selection in Time Series
      1. Introduction
      2. 10.6 - Planning Versus Forecasting and Manual Overrides
      3. 10.7 - Scenario-Based Forecasting
      4. 10.8 - New Product Forecasting
  16. Chapter 11 - Model Building: Alternative Modeling Approaches
    1. 11.1 - Nonlinear Forecasting Models
      1. 11.1.1 - Nonlinear Modeling Features
      2. 11.1.2 - Forecasting Models Based on Neural Networks
      3. 11.1.3 - Forecasting Models Based on Support Vector Machines
      4. 11.1.4 - Forecasting Models Based on Evolutionary Computation
    2. 11.2 - More Modeling Alternatives
      1. 11.2.1 - Multivariate Models
      2. 11.2.2 - Unobserved Component Models (UCM)
  17. Chapter 12 - An Example of Data Mining for Forecasting
    1. 12.1 - The Business Problem
    2. 12.2 - The Charter
    3. 12.3 - The Mind Map
    4. 12.4 - Data Sources
    5. 12.5 - Data Prep
    6. 12.6 - Exploratory Analysis and Data Preprocessing
    7. 12.7 - X Variable Imputation
    8. 12.8 - Variable Reduction and Selection
    9. 12.9 - Modeling
    10. 12.10 - Summary
  18. Appendix A
  19. Appendix B
  20. References
  21. Index
  22. Accelerate Your SAS Knowledge with SAS Books