Understanding Robust and Exploratory Data Analysis

by Hoaglin, David C.; Mosteller, Frederick; Tukey, John W.

• ISBN13: 9780471384915
• ISBN10: 0471384917
• Edition: 1st
• Format: Paperback
• Copyright: 2000-06-02
• Publisher: Wiley-Interscience

Summary

The Wiley Classics Library consists of selected books that have become recognized classics in their respective fields. With these new unabridged and inexpensiveeditions, Wiley hopes to extend the life of these important works by making themavailable to future generations of mathematicians and scientists.Currently available in the Series:T.W. Anderson The Statistical Analysis of Time SeriesT.S. Arthanari & Yadolah Dodge Mathematical Programming in StatisticsEmil Artin Geometric AlgebraNorman T. J. Bailey The Elements of Stochastic Processes with Applications to the Natural SciencesRobert G. Bartle The Elements of Integration andLebesgue Measure George E. P. Box & Norman R. DraperEvolutionary Operation: A Statistical Method for Process Improvement George E. P. Box & George C. TiaoBayesian Inference in Statistical Analysis R. W. CarterFinite Groups of Lie Type: Conjugacy Classes and Complex CharactersR. W. Carter Simple Groups of Lie TypeWilliam G. Cochran & Gertrude M. Cox Experimental Designs, Second EditionRichard Courant Differential and Integral Calculus, Volume IRichard Courant Differential and Integral Calculus, Volume IIRichard Courant & D. Hilbert Methods of Mathematical Physics, Volume IRichard Courant & D. Hilbert Methods of Mathematical Physics, Volume IID. R. Cox Planning of ExperimentsHarold S. M. Coxeter Introduction to Geometry, Second EditionCharles W. Curtis & Irving Reiner Representation Theory of Finite Groups and Associative AlgebrasCharles W. Curtis & Irving Reiner Methods of Representation Theory with Applications to Finite Groups and Orders, Volume ICharles W. Curtis & Irving Reiner Methods of Representation Theory with Applications to Finite Groups and Orders, Volume IICuthbert Daniel Fitting Equations to Data: Computer Analysis of Multifactor Data, Second EditionBruno de Finetti Theory of Probability, Volume IBruno de Finetti Theory of Probability, Volume 2W. Edwards Deming Sample Design in Business ResearchAmos de Shalit & Herman Feshbach Theoretical Nuclear Physics, Volume 1-- Nuclear StructureHarold F. Dodge & Harry G. Romig Sampling Inspection Tables: Single and Double SamplingJ. L. Doob Stochastic ProcessesNelson Dunford & Jacob T. Schwartz Linear Operators, Part One, General TheoryNelson Dunford & Jacob T. Schwartz Linear Operators, Part Two, Spectral Theory--Self Adjoint Operators in Hilbert SpaceNelson Dunford & Jacob T. Schwartz Linear Operators, Part Three, Spectral OperatorsRegina C. Elandt-Johnson & Norman L. Johnson Survival Models and Data AnalysisHerman Feshbach Theoretical Nuclear Physics: Nuclear ReactionsJoseph L. Fleiss Design and Analysis of Clinical ExperimentsBernard Friedman Lectures on Applications-Oriented MathematicsPhillip Griffiths & Joseph Harris Principles of Algebraic GeometryGerald J. Hahn & Samuel S. Shapiro Statistical Models in EngineeringMarshall Hall, Jr. Combinatorial Theory, Second EditionMorris H. Hansen, William N. Hurwitz & William G. Madow Sample Survey Methods and Theory, Volume I--Methods and ApplicationsMorris H. Hansen

Author Biography

DAVID C. HOAGLIN, PhD, is a Fellow of the American Statistical Association.

FREDERICK MOSTELLER, PhD, has been the recipient of several honorary degrees and is a former President of the American Statistical Association.

Table of Contents

Introduction	1	(7)
Stem-And-Leaf Displays	7	(26)
John D. Emerson
David C. Hoaglin
The Basic Display	8	(4)
Some Variations	12	(6)
An Historical Note	18	(1)
Sorting	19	(3)
Background on Number on Number of Lines	22	(7)
Summary	29	(4)
Exercises	31	(2)
Letter Values: A Set of Selected Order Statistics	33	(25)
David C. Hoaglin
Sorting and Ranking	34	(2)
Letter Values	36	(2)
Spreads	38	(3)
Letter-Value Displays	41	(1)
Ideal Letter Values	42	(7)
When the Letter Values Are Equally Spaced	49	(2)
Letter Values as Selected Order Statistics	51	(3)
Summary	54	(4)
Exercises	55	(3)
Boxplots and Batch Comparison	58	(39)
John D. Emerson
Judith Strenio
The Boxplot for a Single Batch	59	(6)
Comparing Batches Using Boxplots	65	(12)
The Spread-versus-Level Plot	77	(10)
Background for Spread-versus-Level Plots	87	(5)
Summary	92	(5)
Exercises	93	(4)
Transforming Data	97	(32)
John D. Emerson
Michael A. Stoto
Power Transformations	98	(6)
Reasons for Transforming	104	(1)
Transforming for Symmetry	105	(6)
Transforming for Other Data Structures	111	(1)
Matched Transformations	112	(9)
Serendipitous Effects of Transformation	121	(3)
When Is Transformation Worthwhile?	124	(2)
Summary	126	(3)
Exercises	127	(2)
Resistant Lines For y Versus x	129	(37)
John D. Emerson
David C. Hoaglin
A Resistant Line from Three Groups	130	(9)
Improving the Iterative Adjustments	139	(8)
Background: Groups and Summaries	147	(7)
Influence and Leverage	154	(4)
Other Alternative Methods	158	(8)
Exercises	163	(3)
Analysis of Two-Way Tables By Medians	166	(45)
John D. Emerson
David C. Hoaglin
Two-Way Tables	167	(2)
Median Polish	169	(7)
Medians versus Means	176	(6)
Least-Absolute-Residuals Fitting	182	(8)
Other Methods of Polishing	190	(2)
Breakdown Bounds for Median Polish	192	(6)
Treatment of Holes	198	(2)
Nonadditivity and the Diagnostic Plot	200	(4)
Summary	204	(7)
Exercises	207	(4)
Examining Residuals	211	(36)
Colin Goodall
Residuals and the Fit	212	(3)
Residuals as Batches	215	(13)
Residuals and Regression	228	(12)
More Sophisticated Plots	240	(1)
Summary	241	(6)
Exercises	243	(4)
Mathematical Aspects of Transformation	247	(36)
John D Emerson
Summary of Four Transformation Plots	248	(2)
The Power Transformations as a Family	250	(7)
A Transformation Plot for Symmetry	257	(3)
Transformation Plot for Equal Spread: The Spread-versus-Level Plot	260	(4)
A Transformation Plot for Straightness	264	(3)
Transformation Plot for the Two-Way Table: The Diagnostic Plot	267	(6)
Historical Notes and References	273	(10)
Exercises	281	(2)
Introduction to More Refined Estimators	283	(14)
David C. Hoaglin
Frederick Mosteller
John W. Tukey
Various Approaches to Estimation	285	(2)
Why the Symmetric Case?	287	(3)
The Dominant Role of Sample Size	290	(1)
Introduction to w- and M-Estimators	291	(3)
Technical Approaches	294	(3)
Comparing Location Estimators: Trimmed Means, Medians, and Trimean	297	(42)
James L. Rosenberger
Miriam Gasko
Main Concepts	298	(8)
Simple L-Estimators	306	(10)
Distributions	316	(8)
Choosing the Robust Estimator	324	(9)
Summary	333	(6)
Exercises	336	(3)
M-Estimators of Location: An Outline of the Theory	339	(65)
Colin Goodall
M-Estimators	341	(6)
Examples	347	(2)
Resistance and Robustness of Efficiency	349	(1)
The Influence Curve	350	(8)
The Shape of ψ	358	(7)
Summary of Criteria	365	(1)
Auxiliary Scale Estimates	365	(3)
Examples of Location Estimators	368	(10)
The Change-of-Value Curve	378	(3)
W-Estimation	381	(4)
Computation and One-Step Estimators	385	(2)
Finite-Sample Variances and Efficiencies	387	(8)
Comparisons of M-Estimators of Location	395	(9)
Exercises	400	(4)
Robust Scale Estimators and Confidence Intervals for Location	404	(29)
Boris Iglewicz
Desirable Properties	405	(2)
Several Simple Scale Estimators	407	(4)
Efficiency of the Simple Scale Estimators	411	(3)
Scale Estimators Based on M-Estimators of Location	414	(4)
Pseudovariances	418	(2)
Robust Confidence Intervals	420	(5)
A Test for Departures from Gaussian Shape	425	(2)
Summary	427	(6)
Exercises	429	(4)
Index	433