You are previewing JMP 10 Modeling and Multivariate Methods.
O'Reilly logo
JMP 10 Modeling and Multivariate Methods

Book Description

JMP 10 Modeling and Multivariate Methods begins by showing you how to take advantage of classic modeling techniques such as linear, nonlinear, and mixed models. The book continues with discussions on neural networking, time series analysis, multivariate techniques, and stepwise regression along with many other JMP modeling and multivariate methods. Examples guide you through each analysis, and statistical references and algorithms are included.

Table of Contents

  1. Cover Page
  2. Title Page
  3. Copyright Page
  4. Contents
  5. Learn About JMP
    1. Book Conventions
    2. JMP Documentation
      1. JMP Documentation Suite
      2. JMP Help
      3. JMP Books by Users
      4. JMPer Cable
    3. Additional Resources for Learning JMP
      1. Tutorials
      2. The JMP Starter Window
      3. Sample Data Tables
      4. Learn about Statistical and JSL Terms
      5. Learn JMP Tips and Tricks
      6. Tooltips
      7. Access Resources on the Web
  6. Introduction to the Fit Model Platform
    1. Launch the Fit Model Platform
    2. Example of the Fit Model Platform
    3. Specify Different Model Types
    4. Construct Model Effects
      1. Add
      2. Cross
      3. Nest
      4. Macros
      5. Attributes
      6. Transformations
    5. Fitting Personalities
    6. Emphasis Options for Standard Least Squares
    7. Model Specification Options
    8. Validity Checks
  7. Fitting Standard Least Squares Models
    1. Example Using Standard Least Squares
    2. The Standard Least Squares Report and Options
    3. Regression Reports
    4. Estimates
    5. Effect Screening
    6. Factor Profiling
    7. Row Diagnostics
    8. Save Columns
    9. Effect Options
    10. Restricted Maximum Likelihood (REML) Method
    11. Method of Moments Results
    12. Singularity Details
    13. Examples with Statistical Details
  8. Fitting Stepwise Regression Models
    1. Overview of Stepwise Regression
    2. Example Using Stepwise Regression
    3. The Stepwise Report
      1. Stepwise Platform Options
      2. Stepwise Regression Control Panel
      3. Current Estimates Report
      4. Step History Report
    4. Models with Crossed, Interaction, or Polynomial Terms
    5. Models with Nominal and Ordinal Terms
    6. Using the Make Model Command for Hierarchical Terms
    7. Performing Logistic Stepwise Regression
    8. The All Possible Models Option
    9. The Model Averaging Option
    10. Using Validation
  9. Fitting Multiple Response Models
    1. Example of a Multiple Response Model
    2. The Manova Report
    3. The Manova Fit Options
    4. Response Specification
      1. Choose Response Options
      2. Custom Test Option
    5. Multivariate Tests
      1. The Extended Multivariate Report
      2. Comparison of Multivariate Tests
      3. Univariate Tests and the Test for Sphericity
    6. Multivariate Model with Repeated Measures
    7. Example of a Compound Multivariate Model
    8. Discriminant Analysis
  10. Fitting Generalized Linear Models
    1. Overview of Generalized Linear Models
    2. The Generalized Linear Model Personality
    3. Examples of Generalized Linear Models
      1. Model Selection and Deviance
    4. Examples
      1. Poisson Regression
      2. Poisson Regression with Offset
      3. Normal Regression, Log Link
    5. Platform Commands
  11. Performing Logistic Regression on Nominal and Ordinal Responses
    1. Introduction to Logistic Models
    2. The Logistic Fit Report
      1. Logistic Plot
      2. Iteration History
      3. Whole Model Test
      4. Lack of Fit Test (Goodness of Fit)
      5. Parameter Estimates
      6. Likelihood Ratio Tests
    3. Logistic Fit Platform Options
      1. Plot Options
      2. Likelihood Ratio Tests
      3. Wald Tests for Effects
      4. Confidence Intervals
      5. Odds Ratios (Nominal Responses Only)
      6. Inverse Prediction
      7. Save Commands
      8. ROC Curve
      9. Lift Curve
      10. Confusion Matrix
      11. Profiler
    4. Validation
    5. Example of a Nominal Logistic Model
    6. Example of an Ordinal Logistic Model
    7. Example of a Quadratic Ordinal Logistic Model
    8. Stacking Counts in Multiple Columns
  12. Analyzing Screening Designs
    1. Overview of the Screening Platform
      1. When to Use the Screening Platform
      2. Comparing Screening and Fit Model
    2. Launch the Screening Platform
    3. The Screening Report
      1. Contrasts
      2. Half Normal Plot
    4. Tips on Using the Platform
    5. Additional Examples
      1. Analyzing a Plackett-Burman Design
      2. Analyzing a Supersaturated Design
    6. Statistical Details
  13. Performing Nonlinear Regression
    1. Introduction to the Nonlinear Platform
    2. Example of Nonlinear Fitting
    3. Launch the Nonlinear Platform
    4. Nonlinear Fitting with Fit Curve
      1. Fit Curve Models and Options
      2. Fit Curve Report
      3. Model Options
    5. Fit a Custom Model
      1. Create a Model Column
      2. Nonlinear Fit Report
      3. Nonlinear Fit Options
      4. Use the Model Library
    6. Additional Examples
      1. Maximum Likelihood: Logistic Regression
      2. Probit Model with Binomial Errors: Numerical Derivatives
      3. Poisson Loss Function
    7. Statistical Details
      1. Profile Likelihood Confidence Limits
      2. How Custom Loss Functions Work
      3. Notes Concerning Derivatives
      4. Notes on Effective Nonlinear Modeling
  14. Creating Neural Networks
    1. Overview of Neural Networks
    2. Launch the Neural Platform
      1. The Neural Launch Window
      2. The Model Launch
    3. Model Reports
      1. Training and Validation Measures of Fit
      2. Confusion Statistics
    4. Model Options
    5. Example of a Neural Network
  15. Modeling Relationships With Gaussian Processes
    1. Launching the Platform
    2. The Gaussian Process Report
      1. Actual by Predicted Plot
      2. Model Report
      3. Marginal Model Plots
    3. Platform Options
    4. Borehole Hypercube Example
  16. Fitting Dispersion Effects with the Loglinear Variance Model
    1. Overview of the Loglinear Variance Model
      1. Model Specification
      2. Notes
    2. Example Using Loglinear Variance
    3. The Loglinear Report
    4. Loglinear Platform Options
      1. Save Columns
      2. Row Diagnostics
    5. Examining the Residuals
    6. Profiling the Fitted Model
      1. Example of Profiling the Fitted Model
  17. Recursively Partitioning Data
    1. Introduction to Partitioning
    2. Launching the Partition Platform
    3. Partition Method
      1. Decision Tree
      2. Bootstrap Forest
      3. Boosted Tree
    4. Validation
    5. Graphs for Goodness of Fit
      1. Actual by Predicted Plot
      2. ROC Curve
      3. Lift Curves
    6. Missing Values
    7. Example
      1. Decision Tree
      2. Bootstrap Forest
      3. Boosted Tree
      4. Compare Methods
    8. Statistical Details
  18. Performing Time Series Analysis
    1. Launch the Platform
    2. Time Series Commands
      1. Graph
      2. Autocorrelation
      3. Partial Autocorrelation
      4. Variogram
      5. AR Coefficients
      6. Spectral Density
      7. Save Spectral Density
      8. Number of Forecast Periods
      9. Difference
    3. Modeling Reports
      1. Model Comparison Table
      2. Model Summary Table
      3. Parameter Estimates Table
      4. Forecast Plot
      5. Residuals
      6. Iteration History
      7. Model Report Options
    4. ARIMA Model
    5. Seasonal ARIMA
    6. ARIMA Model Group
    7. Transfer Functions
      1. Report and Menu Structure
      2. Diagnostics
      3. Model Building
      4. Transfer Function Model
      5. Model Reports
      6. Model Comparison Table
      7. Fitting Notes
    8. Smoothing Models
  19. Performing Categorical Response Analysis
    1. The Categorical Platform
    2. Launching the Platform
    3. Failure Rate Examples
      1. Response Frequencies
      2. Indicator Group
      3. Multiple Delimited
      4. Multiple Response By ID
      5. Multiple Response
    4. Categorical Reports
      1. Report Content
      2. Report Format
      3. Statistical Commands
      4. Save Tables
  20. Performing Choice Modeling
    1. Introduction to Choice Modeling
      1. Choice Statistical Model
    2. Product Design Engineering
    3. Data for the Choice Platform
    4. Example: Pizza Choice
    5. Launch the Choice Platform and Select Data Sets
      1. Choice Model Output
      2. Subject Effects
      3. Utility Grid Optimization
    6. Platform Options
    7. Example: Valuing Trade-offs
    8. One-Table Analysis
      1. Example: One-Table Pizza Data
    9. Segmentation
    10. Special Data Rules
      1. Default choice set
      2. Subject Data with Response Data
      3. Logistic Regression
    11. Transforming Data
      1. Transforming Data to Two Analysis Tables
      2. Transforming Data to One Analysis Table
    12. Logistic Regression for Matched Case-Control Studies
  21. Correlations and Multivariate Techniques
    1. Launch the Multivariate Platform
      1. Estimation Methods
    2. The Multivariate Report
    3. Multivariate Platform Options
      1. Nonparametric Correlations
      2. Scatterplot Matrix
      3. Outlier Analysis
      4. Item Reliability
      5. Impute Missing Data
    4. Examples
      1. Example of Item Reliability
    5. Computations and Statistical Details
      1. Estimation Methods
      2. Pearson Product-Moment Correlation
      3. Nonparametric Measures of Association
      4. Inverse Correlation Matrix
      5. Distance Measures
      6. Cronbach’s α
  22. Clustering Data
    1. Introduction to Clustering Methods
    2. The Cluster Launch Dialog
    3. Hierarchical Clustering
      1. Hierarchical Cluster Options
      2. Technical Details for Hierarchical Clustering
    4. K-Means Clustering
      1. K-Means Control Panel
      2. K-Means Report
    5. Normal Mixtures
      1. Robust Normal Mixtures
      2. Platform Options
      3. Details of the Estimation Process
    6. Self Organizing Maps
  23. Analyzing Principal Components and Reducing Dimensionality
    1. Principal Components
    2. Launch the Platform
    3. Report
    4. Platform Options
      1. Factor Analysis
  24. Performing Discriminant Analysis
    1. Introduction
    2. Discriminating Groups
      1. Discriminant Method
      2. Stepwise Selection
      3. Canonical Plot
      4. Discriminant Scores
    3. Commands and Options
    4. Validation
  25. Fitting Partial Least Squares Models
    1. Overview of the Partial Least Squares Platform
    2. Example of Partial Least Squares
    3. Launch the Partial Least Squares Platform
      1. Launch through Multivariate Methods
      2. Launching through Fit Model
      3. Centering and Scaling
      4. Impute Missing Data
    4. Model Launch Control Panel
    5. The Partial Least Squares Report
    6. Model Fit Options
    7. Partial Least Squares Options
    8. Statistical Details
  26. Scoring Tests Using Item Response Theory
    1. Introduction to Item Response Theory
    2. Launching the Platform
    3. Item Analysis Output
      1. Characteristic Curves
      2. Information Curves
      3. Dual Plots
    4. Platform Commands
    5. Technical Details
  27. Plotting Surfaces
    1. Surface Plots
    2. Launching the Platform
      1. Plotting a Single Mathematical Function
      2. Plotting Points Only
      3. Plotting a Formula from a Column
      4. Isosurfaces
    3. The Surface Plot Control Panel
      1. Appearance Controls
      2. Independent Variables
      3. Dependent Variables
    4. Plot Controls and Settings
      1. Rotate
      2. Axis Settings
      3. Lights
      4. Sheet or Surface Properties
      5. Other Properties and Commands
    5. Keyboard Shortcuts
  28. Visualizing, Optimizing, and Simulating Response Surfaces
    1. Introduction to Profiling
    2. The Profiler
      1. Interpreting the Profiles
      2. Profiler Options
      3. Desirability Profiling and Optimization
    3. Special Profiler Topics
      1. Propagation of Error Bars
      2. Customized Desirability Functions
      3. Mixture Designs
      4. Expanding Intermediate Formulas
      5. Linear Constraints
    4. Contour Profiler
    5. Mixture Profiler
    6. Surface Profiler
    7. The Custom Profiler
    8. The Simulator
      1. Specifying Factors
      2. Specifying the Response
      3. Run the Simulation
      4. The Simulator Menu
      5. Using Specification Limits
      6. Simulating General Formulas
      7. The Defect Profiler
    9. Noise Factors (Robust Engineering)
    10. Profiling Models Stored in Excel
      1. Running the JMP Profiler
      2. Example of an Excel Model
      3. Using the Excel Profiler From JMP
    11. Fit Group
    12. Statistical Details
  29. Comparing Model Performance
    1. Example of Model Comparison
    2. Launch the Model Comparison Platform
    3. The Model Comparison Report
    4. Model Comparison Platform Options
    5. Additional Example of Model Comparison
  30. References
  31. Statistical Details
    1. The Response Models
      1. Continuous Responses
      2. Nominal Responses
      3. Ordinal Responses
    2. The Factor Models
      1. Continuous Factors
      2. Nominal Factors
      3. Ordinal Factors
    3. The Usual Assumptions
      1. Assumed Model
      2. Relative Significance
      3. Multiple Inferences
      4. Validity Assessment
      5. Alternative Methods
    4. Key Statistical Concepts
      1. Uncertainty, a Unifying Concept
      2. The Two Basic Fitting Machines
    5. Leverage Plot Details
    6. Multivariate Details
      1. Multivariate Tests
      2. Approximate F-Test
      3. Canonical Details
      4. Discriminant Analysis
    7. Power Calculations
      1. Computations for the LSV
      2. Computations for the LSN
      3. Computations for the Power
      4. Computations for Adjusted Power
    8. Inverse Prediction with Confidence Limits
    9. Details of Random Effects
  32. Index