[PDF] analytical bias
[PDF] analytical chemistry calculation questions and answers
[PDF] analytical chemistry calculations pdf
[PDF] analytical chemistry pdf
[PDF] analytical chemistry practice problems with solutions
[PDF] analytical chemistry problems and solutions pdf
[PDF] analytical concentration definition
[PDF] analytical concentration vs. equilibrium concentration
[PDF] analytical reports for skeptical audiences
[PDF] analytics as a service companies
[PDF] analytics as a service examples
[PDF] analytics as a service ppt
[PDF] analytics documentation
[PDF] analytics salary survey 2019
[PDF] analytics strategy ppt
Walter Gautschi
Numerical Analysis
Second Edition
Walter Gautschi
Department of Computer Sciences
Purdue University
250 N. University Street
West Lafayette, IN 47907-2066
wgautschi@purdue.edu
ISBN 978-0-8176-8258-3 e-ISBN 978-0-8176-8259-0
DOI 10.1007/978-0-8176-8259-0
Springer New York Dordrecht Heidelberg London
Library of Congress Control Number: 2011941359
Mathematics Subject Classification (2010): 65-01, 65D05, 65D07, 65D10, 65D25, 65D30, 65D32,
65H04, 65H05, 65H10, 65L04, 65L05, 65L06, 65L10
c ?Springer Science+Business Media, LLC 1997, 2012
All rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer ScienceCBusiness Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in
connection with any form of information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology nowknown or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are
not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject
to proprietary rights.
Printed on acid-free paper
www.birkhauser-science.com TO ERIKA
Preface to the Second Edition
In this second edition, the outline of chapters and sections has been preserved. The subtitle ÒAn IntroductionÓ,as suggested by several reviewers, has been deleted. The content, however, is brought up to date, both in the text and in the notes. Many passages in the text have been either corrected or improved. Some biographical notes have been added as well as a few exercises and computer assignments. The typographical appearance has also been improved by printing vectors and matrices consistently in boldface types. With regard to computer language in illustrations and exercises, we now adopt uniformly Matlab. For readers not familiar with Matlab, there are a number of introductory texts available, some, like Moler [2004], Otto and Denier [2005], Stanoyevitch [2005] that combine Matlab with numerical computing, others, like Knight [2000], Higham and Higham [2005], Hunt, Lipsman and Rosenberg [2006], and Driscoll [2009], more exclusively focused on Matlab. Themajornovelty,however,is acompleteset ofdetailedsolutionstoallexercises and machine assignments. The solution manual is available to instructors upon request at the publisherÕs websitehttp://www.birkhauser-science.com/978-0-8176-
8258-3
. Selected solutions are also included in the text to give students an idea of what is expected. The bibliographyhas been expanded to reßect technical advances in the Þeld and to include references to new books and expository accounts. As a result, the text has undergone an expansion in size of about 20%.
West Lafayette, Indiana Walter Gautschi
November 2011
vii
Preface to the First Edition
The book is designed for use in a graduate program in Numerical Analysis that is structured so as to include a basic introductory course and subsequent more specialized courses. The latter are envisaged to cover such topics as numerical linear algebra, the numerical solution of ordinary and partial differential equations, and perhaps additional topics related to complex analysis, to multidimensional analysis, in particularoptimization,and to functionalanalysis and related functional equations. Viewed in this context, the Þrst four chapters of our book could serve as a text for the basic introductory course, and the remaining three chapters (which indeed are at a distinctly higher level) could provide a text for an advanced course on the numerical solution of ordinary differential equations. In a sense, therefore, the book breaks with tradition in that itdoes no longer attempt to deal with all major topics of numerical mathematics. It is felt by the author that some of the current subdisciplines, particularly those dealing with linear algebra and partial differential equations, have developed into major Þelds of study that have attained a degree of autonomy and identity that justiÞes their treatment in separate books and separate courses on the graduate level. The term ÒNumerical AnalysisÓ as used in this book, therefore, is to be taken in the narrow sense of the numerical analogue of Mathematical Analysis, comprising such topics as machine arithmetic, the approximationof functions,approximatedifferentiationand integration,and the approximate solution of nonlinear equations and of ordinary differential equations. What is being covered, on the other hand, is done so with a view toward stressing basic principles and maintaining simplicity and student-friendliness as far as possible. In this sense, the book is ÒAn IntroductionÓ. Topics that, even though important and of current interest, require a level of technicality that transcends the boundsof simplicity striven for, are referencedin detailed bibliographicnotes at the end of each chapter. It is hoped, in this way,to place the material treated in proper contextand to help, indeed encourage,the reader to pursue advancedmodern topics in more depth. A signiÞcant feature of the book is the large collection of exercises that are designed to help the student develop problem-solving skills and to provide interesting extensions of topics treated in the text. Particular attention is given to ix xPreface to the First Edition machine assignments, where the studentis encouraged to implement numerical techniques on the computer and to make use of modern software packages. The author has taught the basic introductory course and the advanced course on ordinary differential equations regularly at Purdue University for the last 30 years or so. The former, typically, was offered both in the fall and spring semesters, to a mixed audience consisting of graduate (and some good undergraduate) students in mathematics, computer science, and engineering,while the latter was taught only in the fall, to a smaller but also mixed audience. Written notes began to materialize in the 1970s, when the authortaught the basic course repeatedlyin summer courses on Mathematics held in Perugia, Italy. Indeed, for some time, these notes existed only in the Italian language. Over the years, th ey were progressively expanded, updated, andtransposedinto English,andalongwith that,notesforthe advancedcoursewere developed. This, briefly, is how the present book evolved. A long gestation period such as this, of course, is not without dangers, the most notable one being a tendency for the material to become dated. The author tried to counteract this by constantly updating and revising the notes, adding newer developments when deemed appropriate. There are, however, benefits as well: over time, one develops a sense for what is likely to stand the test of time and what may only be of temporaryinterest, and one selects and deletes accordingly.Another benefit is the steady accumulation of exercises and the opportunity to have them tested on a large and diverse student population. The purpose of academic teaching, in the author's view, is twofold: to transmit knowledge, and, perhaps more important, to kindle interest and even enthusiasm in the student. Accordingly, the author did not strive for comprehensiveness - even within the boundaries delineated - but rather tried to concentrate on what is essential, interesting and intellectuallypleasing, and teachable. In line with this, an attempt has been made to keep the text uncluttered with numerical examples and otherillustrative material.Being well aware,however,that masteryof a subjectdoes not come from studying alone but from active participation, the author provided many exercises, including machine projects. Attributions of results to specific authors and citations to the literature have been deliberately omitted from the body of the text. Each chapter, as already mentioned, has a set of appended notes that help the reader to pursue related topics in more depth and to consult the specialized literature. It is here where attributions and historical remarks are made, and where citations to the literature - both textbook and research - appear. The main text is preceded by a prologue, which is intended to place the book in proper perspective. In addition to othertextbooks on the subject, and information on software, it gives a detailed list of topics not treated in this book, but definitely belonging to the vast area of computational mathematics, and it provides ample references to relevant texts. A list of numerical analysis journals is also included. The reader is expected to have a good background in calculus and advanced calculus. Some passages of the text require a modest degree of acquaintance with linearalgebra,complexanalysis, ordifferentialequations.Thesepassages, however, can easily be skipped, without loss of continuity, by a student who is not familiar with these subjects.
Preface to the First Editionxi
It is a pleasure to thank the publisher for showing interest in this book and cooperating in producing it. The author is also grateful to Soren Jensen and Manil Suri, who taught from this text, and to an anonymous reader; they all made many helpful suggestions on improving the presentation. He is particularly indebted to Prof. Jensen for substantially helping in preparing the exercises to Chap. 7. The author further acknowledges assistance from Carl de Boor in preparing the notes to Chap. 2 and to Werner C. Rheinboldt for helping with the notes to Chap. 4. Last but not least, he owes a measure of gratitude to Connie Wilson for typing a preliminary version of the text and to Adam Hammer for assisting the author with the more intricate aspects of LaTeX.
West Lafayette, Indiana Walter Gautschi
January 1997
Contents
Prologue.......................................................................... xix P1 Overview.............................................................. xix P2 Numerical Analysis Software........................................ xxi P3 Textbooks and Monographs.......................................... xxi P3.1 Selected Textbooks on Numerical Analysis................. xxi P3.2 Monographsand Books on Specialized Topics............. xxiii P4 Journals................................................................xxvi
1 Machine Arithmetic and Related Matters............................... 1
1.1 Real Numbers, Machine Numbers, and Rounding.................. 2
1.1.1 Real Numbers................................................. 2
1.1.2 Machine Numbers............................................ 3
1.1.3 Rounding..................................................... 5
1.2 Machine Arithmetic.................................................. 7
1.2.1 A Model of Machine Arithmetic............................ 7
1.2.2 Error Propagation in Arithmetic Operations:
Cancellation Error............................................ 8
1.3 The Condition of a Problem.......................................... 11
1.3.1 Condition Numbers.......................................... 13
1.3.2 Examples...................................................... 16
1.4 The Condition of an Algorithm...................................... 24
1.5 Computer Solution of a Problem; Overall Error.................... 27
1.6 Notes to Chapter 1.................................................... 28
Exercises and Machine Assignments to Chapter 1......................... 31 Exercises...................................................................... 31 Machine Assignments........................................................ 39 Selected Solutions to Exercises.............................................. 44 Selected Solutions to Machine Assignments................................ 48
2 Approximation and Interpolation......................................... 55
2.1 Least Squares Approximation........................................ 59
2.1.1 Inner Products................................................ 59
2.1.2 The Normal Equations....................................... 61
xiii xivContents
2.1.3 Least Squares Error; Convergence........................... 64
2.1.4 Examples of Orthogonal Systems........................... 67
2.2 Polynomial Interpolation............................................. 73
2.2.1 Lagrange Interpolation Formula: Interpolation Operator... 74
2.2.2 Interpolation Error............................................ 77
2.2.3 Convergence.................................................. 81
2.2.4 Chebyshev Polynomials and Nodes......................... 86
2.2.5 Barycentric Formula......................................... 91
2.2.6 Newton's Formula............................................ 93
2.2.7 Hermite Interpolation........................................ 97
2.2.8 Inverse Interpolation......................................... 100
2.3 Approximation and Interpolation by Spline Functions............. 101
2.3.1 Interpolation by Piecewise Linear Functions............... 102
2.3.2 A Basis forS
01 .?/............................................ 104
2.3.3 Least Squares Approximation............................... 106
2.3.4 Interpolation by Cubic Splines.............................. 107
quotesdbs_dbs19.pdfusesText_25
[PDF] An Introduction to Numerical Methods and Analysis