Handbook of Data Visualization (eBook)

Table of Contents 5
List of Contributors 8
Part I Data Visualization 13
Introduction 14
Computational Statistics and Data Visualization 15
The Chapters 17
Outlook 23
Principles 24
Part II Principles 24
A Brief History of Data Visualization 25
Introduction 26
Milestones Tour 27
Statistical Historiography 52
Final Thoughts 58
References 59
Good Graphics? 67
Introduction 68
Background 70
Presentation ( What toWhom, How andWhy) 72
Scientific Design Choices in Data Visualization 73
Higher- dimensional Displays and Special Structures 80
Practical Advice 86
And Finally 87
References 87
Static Graphics 89
Complete Plots 91
Customization 94
Extensibility 102
Other Issues 108
Summary 110
References 110
Data Visualization Through Their Graph Representations 112
Introduction 113
Data and Graphs 113
Graph Layout Techniques 115
Discussion and Concluding Remarks 127
References 127
Graph-theoretic Graphics 130
Introduction 131
Definitions 131
Graph Drawing 133
Geometric Graphs 145
Graph-theoretic Analytics 152
References 158
High- dimensional Data Visualization 160
Introduction 161
Mosaic Plots 162
Trellis Displays 166
Parallel Coordinate Plots 173
Projection Pursuit and the Grand Tour 181
Recommendations 184
References 186
Multivariate Data Glyphs: Principles and Practice 188
Introduction 189
Data 189
Mappings 190
Examples of Existing Glyphs 191
Biases in GlyphMappings 192
Ordering of Data Dimensions/Variables 193
Glyph Layout Options 197
Evaluation 200
Summary 204
References 205
Linked Views for Visual Exploration 208
Visual Exploration by Linked Views 209
Theoretical Structures for Linked Views 212
Visualization Techniques for Linked Views 218
Software 222
Conclusion 223
References 223
Linked Data Views 225
Motivation: Why Use Linked Views? 226
The Linked Views Paradigm 229
Brushing ScatterplotMatrices and Other Nonaggregated Views 232
Generalizing to Aggregated Views 235
Distance-based Linking 239
Linking fromMultiple Views 240
Linking to Domain-specific Views 243
Summary 246
Data Used in This Chapter 247
References 248
Visualizing Trees and Forests 250
Introduction 251
Individual Trees 251
Visualizing Forests 263
Conclusion 269
References 271
Methodologies 272
Part III Methodologies 272
Interactive Linked Micromap Plots for the Display of Geographically Referenced Statistical Data 273
Introduction 274
AMotivational Example 278
Design Issues and Variations on StaticMicromaps 280
Web-based Applications of LM Plots 282
Constructing LM Plots 289
Discussion 294
References 297
Grand Tours, Projection Pursuit Guided Tours, andManual Controls 301
Introductory Notes 302
Tours 307
Using Tours with Numerical Methods 316
End Notes 318
References 318
Multidimensional Scaling 321
Proximity Data 322
Metric MDS 325
Non-metric MDS 328
Example: Shakespeare Keywords 331
Procrustes Analysis 336
Unidimensional Scaling 337
INDSCAL 339
Correspondence Analysis and Reciprocal Averaging 344
Large Data Sets and Other Numerical Approaches 347
References 351
Huge Multidimensional Data Visualization: Back to the Virtue of Principal Coordinates and Dendrograms in the New Computer Age 354
Introduction 356
The Geometric Approach to the Statistical Analysis 357
Factorial Analysis 360
Distance Visualization in 365
Principal AxisMethods and Classification: aUnifiedView 370
Computational Issues 371
Factorial Plans and Dendrograms: the Challenge for Visualization 376
An Application: the Survey of Italian Household Income andWealth 382
Conclusion and Perspectives 388
References 390
Multivariate Visualization by Density Estimation 393
Univariate Density Estimates 394
Bivariate Density Estimates 405
Higher- dimensional Density Estimates 410
References 415
Structured Sets of Graphs 418
Introduction 420
Cartesian Products and the Trellis Paradigm 420
ScatterplotMatrices: splomandxysplom 422
Regression Diagnostic Plots 432
Analysis of Covariance Plots 434
Interaction Plots 437
Boxplots 442
Graphical Display of Incidence and Relative Risk 445
Summary 447
File Name Conventions 447
References 447
Regression by Parts: Fitting Visually Interpretable Models with GUIDE 449
Introduction 450
Boston Housing Data – Effects of Collinearity 451
Extension to GUIDE 455
Mussels – Categorical Predictors and SIR 457
Crash Tests – Outlier Detection Under Confounding 461
Car Insurance Rates – Poisson Regression 467
Conclusion 470
References 471
StructuralAdaptiveSmoothing by Propagation – Separation Methods 472
Nonparametric Regression 473
Structural Adaptation 476
An Illustrative Univariate Example 479
Examples and Applications 481
Concluding Remarks 490
References 492
Smoothing Techniques for Visualisation 494
Introduction 495
Smoothing in One Dimension 497
Smoothing in Two Dimensions 503
Additive Models 508
Discussion 512
References 513
Data Visualization via KernelMachines 540
Introduction 541
Kernel Machines in the Framework of an RKHS 542
Kernel Principal Component Analysis 544
Kernel Canonical Correlation Analysis 552
Kernel Cluster Analysis 555
References 559
Visualizing Cluster Analysis and FiniteMixtureModels 561
Introduction 562
Hierarchical Cluster Analysis 564
Partitioning Cluster Analysis 568
Model-Based Clustering 580
Summary 586
References 586
Visualizing Contingency Tables 588
Introduction 589
Two- Way Tables 590
Using Colors for Residual-Based Shadings 597
Selected Methods for Multiway Tables 605
Conclusion 613
References 613
Mosaic Plots and Their Variants 616
Definition and Construction 618
Interpreting Mosaic Plots 621
Variants 626
RelatedWork and Generalization 634
Implementations 639
References 640
Parallel Coordinates: Visualization, Exploration and Classification of High- Dimensional Data 642
Introduction 643
Exploratory Data Analysis with 647
coords 647
Classification 663
Visual and Computational Models 667
Parallel Coordinates: Quick Overview 670
Future 675
References 677
Matrix Visualization 680
Introduction 681
RelatedWorks 681
The Basic Principles of Matrix Visualization 682
Generalization and Flexibility 689
An Example 692
Comparison with Other Graphical Techniques 696
Matrix Visualization of Binary Data 699
OtherModules and Extensions ofMV 703
Conclusion 704
References 705
Visualization in Bayesian Data Analysis 708
Introduction 709
Using Visualization to Understand and Check Models 711
Example: A HierarchicalModel of Structure in Social Networks 715
Challenges Associated with the Graphical Display of Bayesian Inferences 721
Summary 721
References 722
Programming Statistical Data Visualization in the Java Language 724
Introduction 725
Basics of Statistical Graphics Libraries and Java Programming 726
Design and Implementation of a Java Graphics Library 734
Concluding Remarks 752
References 754
Web-Based Statistical Graphics using XML Technologies 756
Introduction 757
XML-Based Vector Graphics Formats 758
SVG 764
X3D 770
Applications 776
References 787
Selected Applications 789
Part IV Selected Applications 789
Visualization for Genetic Network Reconstruction 790
Introduction 791
Visualization for Data Preprocessing 791
Visualization for Genetic Network Reconstruction 794
References 806
Reconstruction, Visualization and Analysis ofMedical Images 809
Introduction 810
PET Images 811
Ultrasound Images 815
Magnetic Resonance Images 818
Conclusion and Discussion 822
References 824
Exploratory Graphics of a Financial Dataset 827
Introduction 828
Description of the Data 829
First Graphics 830
Outliers 833
Scatterplots 837
Mosaic Plots 839
Initial Comparisons Between Bankrupt Companies 840
Investigating Bigger Companies 844
Summary 847
Software 848
References 848
Graphical Data Representation in Bankruptcy Analysis 849
Company RatingMethodology 850
The SVM Approach 853
Company Score Evaluation 856
Variable Selection 856
Conversion of Scores into PDs 861
Colour Coding 863
Conclusion 867
References 867
Visualizing Functional Data with an Application to eBay’s Online Auctions 869
Introduction 870
Online Auction Data fromeBay 872
Visualization at the Object Recovery Stage 873
Visualizing Functional Observations 878
Interactive Information Visualization of Functional and Cross- sectional Information via TimeSearcher 886
Further Challenges and Future Directions 892
References 893
Visualization Tools for Insurance Risk Processes 895
Introduction 896
Software 898
Fitting Loss andWaiting Time Distributions 898
Risk Process and its Visualization 908
References 916
Subject Index 917