Dodge Y. (ed.) Statistical Data Analysis Based on the L1-Norm and Related Methods

Basel: Birkhäuser, 2002. — 446 p.This volume contains a selection of invited papers, presented to the fourth In­ Statistical Analysis Based on the L1-Norm and Related ternational Conference on Methods, held in Neuchatel, Switzerland, from August 4-9, 2002. Organized jointly by the University of Illinois at Chicago (Gib Bassett), the Rutgers University (Regina Liu and Yehuda Vardi) and the University of Neuchatel (Yadolah Dodge), the conference brought together experts whose research deals with theory and ap­ plications involving the L1-Norm. The conference included invited and contributed talks as well as a tutorial on Quantile Regression. This volume includes 36 refereed invited papers under seven headings. Part one deals with Quantiles in all their forms and shapes. It includes papers on quantile functions in non-parametric multivariate analysis, and empirical applications of quantile regression. Much of the development in this direction follows from the fundamental paper by Koenker and Bassett in 1978. Financial and Time Series A nalysis follows the section on quantiles. Part three concerns Estimation, Testing and Characterization. Part four, Deep in the Data, deals with issues related to data depth. Part five addresses Classification questions. The problem of Density Estimation and Image Processing is discussed in Part six, and finally Part seven presents two environmental applications. The contributions represent clear evidence of important research involving theo­ retical issues and applications associated with the L1-Norm. It is my hope that the articles contained in this volume and its predecessors, published in 1987, 1992, and 1997, will stimulate interest among researchers.Front Matter
Front Matter
Quantile Functions and Spread for Multivariate Distributions
A New Definition of Multivariate M-quantiles
A Depth Function and a Scale Curve Based on Spatial Quantiles
Sample Quantiles for Locally Dependent Processes
What are the Limiting Distributions of Quantile Estimators?
New Selection Indices for University Admissions: A Quantile Approach
Exploring Transition Data through Quantile Regression Methods: An Application to U.S. Unemployment Duration
Front Matter
How to be Pessimistic: Choquet Risk and Portfolio Optimization
Expected Shortfall and Beyond
Credit Scoring Using Binary Quantile Regression
Prediction of 0–1-Events for Short- and Long-memory Time Series
Nonparametric Quantile Regression With Applications to Financial Time Series
An Algorithm for Optimal Bandwidth Selection for Smooth Nonparametric Quantile Estimation
Front Matter
Orthogonal L 1 -norm Estimation
L 1 -Derivatives, Score Functions and Tests
Optimal Bias Robust M—estimates of Regression
Robust Bootstrap for S-estimators of Multivariate Regression
M-tests for Detection of Structural Changes in Regression
Change Point Detection Based on Empirical Quantiles
A Class of Probability Metrics and its Statistical Applications
Front Matter
Whose Hare and Whose Tortoise
Sign and Rank Covariance Matrices: Statistical Properties and Application to Principal Components Analysis
Multivariate Signed Ranks: Randles’ Interdirections or Tyler’s Angles?
Front Matter
L 1 -Depth, Depth Relative to a Model, and Robust Regression
Perturbation Properties of Depth Regions
Multivariate Trimmed Means Based on Data Depth
Front Matter
Graphs, L 1 -Metrics and Clustering
Classification Based on the Support Vector Machine and on Regression Depth
A Robust Clustering Method and Visualization Tool Based on Data Depth
The Median Extension of Data Analysis Metric Structures
Text Classification for Mining Massive Aviation Inspection Reports
Front Matter
A Comparison Between L 1 Markov Random Field-based and Wavelet-based Estimators
Elastic and Plastic Splines: Some Experimental Comparisons
On the Bitplane Compression of Microarray Images
Front Matter
Overdispersed Regression Models for Air Pollution and Human Health
Atmospheric Pollution and Mortality in São Paulo
Back Matter