skip to content

Contents

Preface v
    
1 Introduction  1
1.1 The Brief  1
1.2 Representing a probability measure  2
1.3 Lift zonoids  4
1.4 Example of lift zonoids  9
1.5 Representing distributions by convex compacts  14
1.6 Ordering distributions  16
1.7 Central regions and data depth 19
1.8 Statistical inference  22
2 Zonoids and lift zonoids  25
  
2.1 Zonotopes and lift zonoids 27
2.1.1 Zonoid of a measure 27
2.1.2 Equivalent definitions of the zonoid of a measure  30
2.1.3 Support function of a zonoid  32
2.1.4 Zonoids as expected random segments 34
2.1.5 Volume of a zonoid  35
2.1.6 Measures with equal zonoids  38
2.2 Lift zonoid of a measure  40
2.2.1 Definition and first properties 40
2.2.2 Lift zonotope  43
2.2.3 Univariate case 43
2.3 Embedding into convex compacts  48
2.3.1 Inclusion of lift zonoids  49
2.3.2 Uniqueness of the representation  50
2.3.3 Lift zonoid metric  51
2.3.4 Linear transformations and projections  52
2.3.5 Lift zonoid of spherical and elliptical distributions 55
2.4 Continuity and approximation  58
2.4.1 Convergence of lift zonoids  59
2.4.2 Monotone approximation of measures  65
2.4.3 Volume of a lift zonid  66
2.5 Limit theorems  67
2.6 Representation of measures by a functional  70
2.6.1 Statistical representations  74
2.6.2 Lift zonoids and the empirical process  76
2.7 Notes 77
3 Central Regions  79
  
3.1 Zonoid trimmed regions 81
3.2 Properties  84
3.3 Univariate central regions  85
3.4 Examples of zonoid trimmed regions  88
3.5 Notions of central regions  93
3.6 Continuity and law of large numbers  96
3.7 Further properties 97
3.8 Trimming empirical measures  100
3.9 Computation of zonoid trimmed regions  102
3.10 Notes  103
4 Data Depth 105
4.1 Zonoid depth  108
4.2 Properties of the zonoid depth  111
4.3 Different notions of data depth  115
4.4 Combinatorial invariance  122
4.5 Computation of the zonoid depth  127
4.6 Notes  129
  
5 Inference based on data depth (by Rainer Dyckerhoff)  131
  
5.1 General notion of data depth  132
5.2 Two-sample depth test for scale  134
5.3 Two-sample rank test for location and scale  137
5.4 Classical two-sample tests  139
5.4.1 Box´s M Test  139
5.4.2 Friedman-Rafsky test  140
5.4.3 Hotelling´s T² test  142
5.4.4 Puri-Sen test  143
5.5 A new Wilcoxon distance test  145
5.6 Power comparison  147
5.7 Notes 161
6 Depth of hyperplanes  163
  
6.1 Depth of a hyperplane and MHD of a sample  164
6.2 Properties of MHD and majority depth  166
6.3 Combinatorial invariance  169
6.4 Measuring combinatorial dispersion  171
6.5 MHD statistics 172
6.6 Significance tests and their power  172
6.7 Notes  177
  
7 Volume statistics  179
  
7.1 Univariate Gini Index  180
7.2 Lift zonoid volume  184
7.3 Expected volume of a random convex hull  186
7.4 The multivariate volume-Gini index  189
7.5 Volume statistics in cluster analysis  195
7.6 Measuring dependency  196
7.7 Notes 203
8 Ordering and indices of dispersion 205
8.1 Lift zonoid order 206
8.2 Order of marginals and independence  211
8.3 Order of convolutions  212
8.4 Lift zonoid order vs. convex order  214
8.5 Volume inequalities and random determinants  217
8.6 Increasing, scaled, and centred orders  217
8.7 Properties of dispersion orders  220
8.8 Multivariate indices of dispersion  222
8.9 Notes  226
  
9 Economic disparity and concentration  227
  
9.1 Measuring economic inequality  228
9.2 Inverse Lorenz function (ILF)  230
9.3 Price Lorenz order  236
9.4 Majorizations of absolute endowments  240
9.5 Other inequality orderings 243
9.6 Measuring industrial concentration 246
9.7 Multivariate concentration function  250
9.8 Multivariate concentration indices  253
9.9 Notes  255
  
Appendix A: Basic Notions  257
Appendix B: Lift zonoids of bivariate normals  263
Bibliography  272
Index  286