Comprehensive Chemometrics

Romà Tauler; Beata Walczak; Steven D. Brown; Steven D. Brown

BOOK

Comprehensive Chemometrics

Romà Tauler | Beata Walczak | Steven D. Brown | Steven D. Brown

(2009)

Additional Information

Book Details

ISBN: 978-0-444-52701-1
Edition
Language: English
Pages: 2896
Subjects: Surgical orthopaedics & fractures
Medicinal chemistry
Analytical chemistry
Organic chemistry
Biochemistry
Chemical biology

Abstract

Designed to serve as the first point of reference on the subject, Comprehensive Chemometrics presents an integrated summary of the present state of chemical and biochemical data analysis and manipulation. The work covers all major areas ranging from statistics to data acquisition, analysis, and applications.

This major reference work provides broad-ranging, validated summaries of the major topics in chemometrics—with chapter introductions and advanced reviews for each area. The level of material is appropriate for graduate students as well as active researchers seeking a ready reference on obtaining and analyzing scientific data.

Features the contributions of leading experts from 21 countries, under the guidance of the Editors-in-Chief and a team of specialist Section Editors: L. Buydens; D. Coomans; P. Van Espen; A. De Juan; J.H. Kalivas; B.K. Lavine; R. Leardi; R. Phan-Tan-Luu; L.A. Sarabia; and J. Trygg
Examines the merits and limitations of each technique through practical examples and extensive visuals: 368 tables and more than 1,300 illustrations (750 in full color)
Integrates coverage of chemical and biological methods, allowing readers to consider and test a range of techniques
Consists of 2,200 pages and more than 90 review articles, making it the most comprehensive work of its kind
Offers print and online purchase options, the latter of which delivers flexibility, accessibility, and usability through the search tools and other productivity-enhancing features of ScienceDirect

"An excellent source of up-to-date information…essential for researchers and others working in the field." --Analytical and Bioanalytical Chemistry

"Contains much to interest the spectroscopist…well worthy of a place in any analytical science library." --Spectroscopy Europe

"Highly recommended…Should find a merited place on bookshelves alongside its predecessors." --Alexey L. Pomerantsev and Oxana Ye. Rodionova, Semenov Institute of Chemical Physics RAS, Russia

Section Title	Page	Action	Price
e9780444527028v1	1
Series page	3
Title page	4
Copyright \rPage	5
Contents of Volume 1\r	6
Contributors to Volume 1	8
Preface	10
Editors in Chief	11
Contents of All Volumes	12
Section Editors	16
1.01 An Introduction to the Theory of Sampling: An EssentialPart of Total Quality Management	20
Symbols	36
1.01.1 Introduction	21
1.01.2 Scope	21
1.01.3 Definitions and Notations	22
1.01.4 Dividing a Complex Problem into its Basic Components	22
1.01.5 Exercises Challenging the Reader	25
1.01.6 The Critical Importance of Sampling Courses	28
1.01.7 The Enemies and Their Link to Geostatistics	29
1.01.8 Large-Scale Variability	29
1.01.9 Conclusions	32
1.01.10 Recommendations	32
References	34
1.02 Quality of Analytical Measurements: StatisticalMethods for Internal Validation	36
1.02.1 Introduction	37
1.02.2 Confidence and Tolerance Intervals	42
1.02.3 Hypothesis Test	50
1.02.4 One-Way Analysis of Variance	64
1.02.5 Statistical Inference and Validation	73
Appendix	84
References	91
1.03 Proficiency Testing in Analytical Chemistry	96
Symbols	96
1.03.1 Overview of Proficiency Testing in Analytical Chemistry	97
1.03.2 z-Scoring	98
1.03.3 Validation of Test Materials	109
1.03.4 Further Information from Proficiency Testing Results	112
References	113
1.04 Statistical Control of Measures and Processes	116
Symbols	116
1.04.1 Introduction: Basics of Process Monitoring	119
1.04.2 Phases in Process Monitoring	121
1.04.3 Shewhart Control Charts	121
1.04.4 CUSUM Control Charts	123
1.04.5 EWMA Control Charts	124
1.04.6 Performance Measures of Control Charts	125
1.04.7 Control Charts for Autocorrelated Processes	127
1.04.8 Integration of SPC and Engineering Process Control	130
1.04.9 Multivariate Control Charts	131
1.04.10 Software	142
Acknowledgments	142
References	143
1.05 Quality of Analytical Measurements: UnivariateRegression	146
Symbols	147
1.05.1 Introduction	147
1.05.2 Linear Regression in Calibration: Elements and Procedure	148
1.05.3 Statistical Validation of a Calibration Model	153
1.05.4 Confidence Intervals and Hypothesis Testing	159
1.05.5 The Design of a Calibration	167
1.05.6 The Capability of Detection, the Decision Limit, and the Capability of Discrimination Computed from a Regression Model	170
1.05.7 Standard Addition Method	172
1.05.8 Weighted Least Squares and Generalized Least Squares	173
1.05.9 Robust Regression in Calibration	178
1.05.10 Errors in Both Variables	182
1.05.11 Final Remark	183
References	184
1.06 Resampling and Testing in Regression Models withEnvironmetrical Applications	190
1.06.1 Introduction to Bootstrap	190
1.06.2 Bootstrap Resampling Methods for Regression	192
1.06.3 Generalized Additive Models	193
1.06.4 Constructing CIs in GAMs	195
1.06.5 Generalized Additive Model with Interactions	198
1.06.6 Bootstrap-Based Methods for Testing Interactions	199
References	203
1.07 Robust and Nonparametric Statistical Methods	208
Symbols	208
1.07.1 Introduction	209
1.07.2 Location and Scale Estimation	209
1.07.3 Correlation and Covariance	214
1.07.4 Regression	216
1.07.5 Investigation Dependence Structures	222
1.07.6 Bibliographic Notes	227
Acknowledgments	228
References	228
1.08 Bayesian Methodology in Statistics	232
Symbols	233
1.08.1 Introduction and Notation	233
1.08.2 Axiomatic Foundations	235
1.08.3 Bayesian Methodology	238
1.08.4 Reference Analysis	247
1.08.5 Inference Summaries	255
1.08.6 Discussion	261
Acknowledgments	262
References	262
1.09 Experimental Design: Introduction	266
1.09.1 Screening	267
1.09.2 Quantitative Study of the Factors	267
1.09.3 Response Surface Methodology	267
1.09.4 Mixtures or Formulations	267
1.09.5 Nonclassical Strategies	268
1.10 Screening Strategies	270
1.10.1 Introduction	270
1.10.2 Screening Saturated Designs	273
1.10.3 Supersaturated Designs	283
1.10.4 Screening Designs at More than Two Levels	295
1.10.5 Multilevel Supersaturated Designs	300
1.10.6 Applications of Supersaturated Designs	310
1.10.7 Composite Samples and Group Screening	311
1.10.8 Sequential Bifurcation	315
References	316
1.11 The Study of Experimental Factors	320
1.11.1 Introduction	321
1.11.2 Factorial Designs	329
1.11.3 Fractional Factorial Designs	348
1.11.4 Concluding Remarks	361
Acknowledgments for the Permissions to Use Copyrighted Material	362
References	362
1.12 Response Surface Methodology	364
1.12.1 Introduction	365
1.12.2 Elements and Notation	365
1.12.3 Optimality of the Variance of the Estimates	369
1.12.4 Some Aspects of the Statistical Validation of the Model	373
1.12.5 Experimental Design for Fitting Response Surfaces	385
1.12.6 Analysis of a Quadratic Response Surface	394
1.12.7 Final Remark	406
References	407
1.13 Experimental Design for Mixture Studies	410
1.13.1 Introduction	410
1.13.2 Mixture Domain	412
1.13.3 Mixture Design for Simplex (-Shaped) Regions	428
1.13.4 Designs for Constrained Mixtures	433
1.13.5 Analysis and Optimization	449
1.13.6 Concluding Remarks	468
References	468
1.14 Nonclassical Experimental Designs	472
1.14.1 Introduction	472
1.14.2 Methodological Approach	474
1.14.3 Criteria	476
1.14.4 Combined Designs	490
1.14.5 Final Remark	513
References	516
1.15 Experimental Designs: Conclusions, Terminology, andSymbols	520
1.15.1 Terminology	521
1.15.2 Symbols	522
1.16 Constrained and Unconstrained Optimization	526
Symbols	526
1.16.1 Introduction	527
1.16.2 Numerical Optimization	528
1.16.3 Optimization in Chemometrics	530
1.16.4 Unconstrained Optimization Methods	531
1.16.5 Globalization Strategies	543
1.16.6 Constrained Optimization Methods	550
1.16.7 Discussion	561
Acknowledgments	562
References	562
1.17 Sequential Optimization Methods	566
Symbols	566
1.17.1 Introduction	568
1.17.2 Sequential Optimization Methods	570
1.17.3 Mixed Sequential–Simultaneous Optimization Methods	588
1.17.4 Conclusions	591
References	591
1.18 Steepest Ascent, Steepest Descent, and GradientMethods	596
Symbols	596
1.18.1 Introduction	596
1.18.2 Method	599
1.18.3 Examples	603
1.18.4 Conclusion	608
References	608
1.19 Multicriteria Decision-Making Methods	610
1.19.1 Introduction	610
1.19.2 Basic Notation	613
1.19.3 Illustrative Example	614
1.19.4 Multi-Criteria Decision Making Methods	615
References	644
1.20 Genetic Algorithms	650
1.20.1 Introduction	650
1.20.2 The Evolution Theory	651
1.20.3 How to Transform the Evolution Theory into an Optimization Technique?	652
1.20.4 The Problem of Coding	652
1.20.5 Steps of the GAs	654
1.20.6 Comments about the Parameters of the GAs	658
1.20.7 Hybrid Algorithms	659
1.20.8 Looking for the Global Maximum: Is it Worthwhile?	660
1.20.9 Applications	660
1.20.10 A Specific Application: Variable Selection in Spectral Data Sets	661
References	670
Index to Volume 1	674
e9780444527028v2	692
Series page	694
Title page	695
Copyright page	696
Contents of Volume 2	697
Contributors to Volume 2	701
Preface	705
Editors in Chief	706
Contents of All Volumes	707
Section Editors	711
2.01 Background Estimation, Denoising, and Preprocessing	715
2.01.1 Introduction	715
2.01.2 Improving the Signal-to-Noise Ratio	716
2.01.3 Variable Shift, Alignment, and Normalization	717
2.01.4 Model-Based Approaches in Preprocessing	719
2.01.5 Conclusions and Discussion	719
References	720
2.02 Denoising and Signal-to-Noise Ratio Enhancement: Classical Filtering	723
Symbols	723
2.02.1 Introduction	723
2.02.2 Nonrecursive Filters	724
2.02.3 Recursive Filters	733
References	737
2.03 Denoising and Signal-to-Noise Ratio Enhancement: Wavelet Transform and Fourier Transform	739
Glossary	739
2.03.1 Introduction	740
2.03.2 Basics of Fourier and Wavelet Analysis	741
2.03.3 Off-Line Denoising	750
2.03.4 On-line Denoising	760
2.03.5 Conclusions	768
References	768
2.04 Denoising and Signal-to-Noise Ratio Enhancement: Derivatives Derivatives	771
Symbols	771
2.04.1 Derivatives	771
References	779
2.05 Denoising and Signal-to-Noise Ratio Enhancement: Splines	781
Symbols	781
2.05.1 Splines	781
References	795
2.06 Variable Shift and Alignment	799
Symbols	799
2.06.1 Introduction	800
2.06.2 Data Reduction	804
2.06.3 Dynamic Programming	804
2.06.4 Alignment Techniques	806
2.06.5 Settings and Applications	819
References	820
2.07 Normalization and Closure	823
Symbols	823
Glossary	824
2.07.1 Introduction	824
2.07.2 Statistical Considerations	826
2.07.3 General Normalization Methods	828
2.07.4 DNA Microarray Normalization	833
2.07.5 Assessment and Quality Control of Normalization Performance	836
2.07.6 Conclusions	838
References	838
2.08 Model Based Preprocessing and Background Elimination: OSC, OPLS, and O2PLS	843
Symbols	843
Glossary	844
2.08.1 Orthogonal Signal Correction	844
2.08.2 Orthogonal Projections to Latent Structures	845
2.08.3 Bidirectional OPLS	847
2.08.4 Kernel-Based Orthogonal Projections to Latent Structures	849
References	850
2.09 Standard Normal Variate, Multiplicative Signal Correction and Extended Multiplicative Signal Correction Preprocessing in Biospectroscopy	853
Symbols	853
2.09.1 Introduction	854
2.09.2 Ideal and Nonideal Spectral Measurements	854
2.09.3 Baseline and Scaling Problems	856
2.09.4 Linearization of Instrument Response	857
2.09.5 Unwanted Instrument-Induced Channel Shifts	858
2.09.6 Random Errors	858
2.09.7 Advanced Baseline and Scaling Problems	859
2.09.8 Formal Description of Extended Multiplicative Signal Correction	862
2.09.9 Preprocessing Examples	864
2.09.10 Conclusions	873
Acknowledgments	873
References	873
2.10 Batch Process Modeling and MSPC	877
2.10.1 Introduction	877
2.10.2 Industrial Examples	879
2.10.3 Analyzing Historical Batch Process Data	881
2.10.4 Online Prediction of Future Trajectories and Final Quality	898
2.10.5 Monitoring (MSPC) of Batch Processes	900
2.10.6 Control and Optimization of Batch Processes	906
2.10.7 Conclusions	908
References	908
2.11 Evaluation of Preprocessing Methods	913
Symbols	913
2.11.1 Introduction	913
2.11.2 Theoretical Background and Methods	914
2.11.3 Example of Evaluation of Preprocessing	915
2.11.4 Discussion and Conclusions	918
References	919
2.12 Linear Soft-Modeling: Introduction	921
2.13 Principal Component Analysis: Concept, Geometrical Interpretation, Mathematical Background, Algorithms, History, Practice	925
Symbols	925
2.13.1 Introduction	925
2.13.2 PCA: Basic Concepts and Master Equation	926
2.13.3 Geometrical Properties of PCA	928
2.13.4 Mathematical Background	931
2.13.5 History	934
2.13.6 Application Example	935
References	938
2.14 Independent Component Analysis	941
2.14.1 Representation of Multivariate Data	941
2.14.2 Blind Source Separation	944
2.14.3 Independent Component Analysis	945
2.14.4 ICA Estimation Principles	948
2.14.5 Validation of Independent Components	950
2.14.6 Evaluation of Algorithmic Performance	951
2.14.7 Tutorial Examples	953
2.14.8 Conclusions	960
Acknowledgments	960
References	960
2.15 Introduction to Multivariate Curve Resolution	963
2.15.1 Key Concepts	963
2.15.2 History	967
2.15.3 Ambiguities	968
2.15.4 Applications	968
References	970
2.16 Two-Way Data Analysis: Evolving Factor Analysis	975
2.16.1 Introduction	975
2.16.2 The Data	975
2.16.3 Abstract Factor Analysis	976
2.16.4 Traditional EFA	978
2.16.5 Fixed Size Moving Window-EFA	980
2.16.6 Exhaustive-EFA	982
2.16.7 EFA of Images, Two-Dimensional-EFA	983
2.16.8 Computational Aspects	985
2.16.9 Using the Results of EFA	986
References	987
2.17 Two-Way Data Analysis: Detection of Purest Variables	989
Symbols	989
Glossary	990
2.17.1 Introduction	990
2.17.2 Key Set Factor Analysis	991
2.17.3 Simple-To-Use Interactive Self-Modeling Mixture Analysis	1001
2.17.4 Orthogonal Projection Approach	1010
2.17.5 Stepwise Maximum Angle Calculation	1012
2.17.6 Conclusion	1017
Acknowledgment	1019
References	1019
2.18 Two-Way Data Analysis: Multivariate Curve Resolution – Noniterative Resolution Methods	1023
2.18.1 Introduction	1023
2.18.2 Evolving Feature and Informative Windows in Two-Way Data	1024
2.18.3 Evolving Factor Analysis	1025
2.18.4 Heuristic Evolving Latent Projections	1026
2.18.5 Window Factor Analysis	1030
2.18.6 Subwindow Factor Analysis	1033
References	1035
2.19 Two-Way Data Analysis: Multivariate Curve Resolution – Iterative Resolution Methods	1039
Iterative Resolution Methods	1039
2.19.1 Introduction	1039
2.19.2 Iterative Methods Based on the Optimization of Transformation Matrices	1044
2.19.3 Methods Based on the Optimization of Concentration Profiles and/or Spectra	1049
2.19.4 Conclusions	1054
References	1055
2.20 Two-Way Data Analysis: Multivariate Curve Resolution – Error in Curve Resolution	1059
2.20.1 Introduction	1059
2.20.2 Characterization of MCR Ambiguities and Error Maps	1061
2.20.3 Estimation of MCR Feasible Band Boundaries	1064
2.20.4 Error Propagation in MCR Solutions	1069
2.20.5 Conclusions	1075
References	1075
2.21 Multiway Data Analysis: Eigenvector-Based Methods	1079
Symbols	1080
2.21.1 Introduction	1081
2.21.2 Notation and Conventions	1083
2.21.3 The Model and Uniqueness Conditions	1083
2.21.4 Background and Current Position of GRAM	1087
2.21.5 Theoretical Properties of GRAM Results	1097
2.21.6 Practical Aspects of GRAM	1102
2.21.7 Software	1108
2.21.8 Applications	1108
2.21.9 Conclusions and Recommendations	1110
Appendix 1: Derivation of SK and WSK Methods	1110
Appendix 2: Derivation of the DTD Method	1112
Appendix 3: The Algebraic Eigenvalue Problem	1113
Appendix 4: Matlab Implementation of SK and WSK Methods	1113
References	1114
2.22 Multilinear Models: Iterative Methods	1125
Symbols	1125
2.22.1 Introduction	1126
2.22.2 PARAFAC Models	1127
2.22.3 Tucker Models	1145
2.22.4 Hybrid Models	1156
Appendix	1160
Vectorization, Matricization, and Other Rearrangements of Multiway Arrays	1162
References	1162
2.23 Multiset Data Analysis: ANOVA Simultaneous Component Analysis and Related Methods	1167
2.23.1 Introduction	1167
2.23.2 Theory	1168
2.23.3 Case Study	1174
2.23.4 Relationship of ASCA to Other Methods	1181
2.23.5 Further Developments	1182
Acknowledgments	1182
References	1183
2.24 Multiset Data Analysis: Extended Multivariate Curve Resolution	1187
2.24.1 Introduction	1187
2.24.2 Extension of Multivariate Curve Resolution to the Simultaneous Analysis of Multiple Data Matrices (Multiway Data and Multiset Data)	1188
2.24.3 Including Hard Modeling Constraints in Multivariate Curve Resolution	1201
2.24.4 Applications	1205
2.24.5 Conclusions	1215
References	1215
2.25 Other Topics in Soft-Modeling: Maximum Likelihood-Based Soft-Modeling Methods	1221
2.25.1 Introduction	1222
2.25.2 Maximum Likelihood Principal Components Analysis	1228
2.25.3 Practical Considerations for MLPCA	1242
2.25.4 Related Techniques	1255
2.25.5 Applications	1257
2.25.6 Conclusions	1270
References	1270
2.26 Unsupervised Data Mining: Introduction	1273
Symbols	1273
2.26.1 Introduction	1273
2.26.2 Data Sets	1274
2.26.3 Uncovering Data Heterogeneity: Cluster Analysis and Data Mapping	1275
2.26.4 Clustering Challenges	1280
2.26.5 Summary	1286
References	1286
2.27 Common Clustering Algorithms	1291
Symbols	1292
2.27.1 Taxonomy of Clustering Algorithms	1292
2.27.2 Partitioning Clustering	1294
2.27.3 Hierarchical Clustering	1312
2.27.4 Hybrid Clustering	1322
2.27.5 Constrained Clustering	1326
References	1328
2.28 Data Mapping: Linear Methods versus Nonlinear Techniques	1333
2.28.1 Introduction	1333
2.28.2 Linear Methods	1334
2.28.3 Nonlinear Methods	1340
2.28.4 Discussion	1344
References	1345
2.29 Density-Based Clustering Methods	1349
2.29.1 Introduction	1349
2.29.2 Clustering Methods Employing the Concept of Data Density	1349
2.29.3 Similarities among Discussed Density-Based Approaches	1359
2.29.4 Advantages of the Density-Based Techniques over the Classical Ones	1359
2.29.5 Applications	1361
References	1366
2.30 Model-Based Clustering	1369
Symbols	1369
2.30.1 Introduction	1370
2.30.2 Definition of Mixture Models	1371
2.30.3 Maximum Likelihood Estimation	1371
2.30.4 Fitting Mixture Models via the EM Algorithm	1372
2.30.5 Choice of Starting Values for the EM Algorithm	1374
2.30.6 Clustering via Normal Mixtures	1375
2.30.7 Spectral Representation of Component-Covariances Matrices	1376
2.30.8 Multivariate t-Distribution	1377
2.30.9 ML Estimation of Mixtures of t-Distributions	1377
2.30.10 Choice of the Number of Components in a Mixture Model	1379
2.30.11 Advantages of Mixture Model-Based Clustering	1380
2.30.12 Factor Analysis Model for Dimension Reduction	1380
2.30.13 Mixtures of Normal Factor Analyzers	1381
2.30.14 Mixtures of t-Factor Analyzers	1385
2.30.15 Available Software	1387
2.30.16 Example	1387
2.30.17 Some Recent Extensions for High-Dimensional Data	1391
2.30.18 Mixed Feature Data	1391
Acknowledgments	1392
References	1393
2.31 Tree-Based Clustering and Extensions	1397
2.31.1 Introduction	1397
2.31.2 Regression Trees for Clustering	1399
2.31.3 Ensemble Co-occurrence Representation	1403
2.31.4 Clustering Co-occurrence Matrices	1409
2.31.5 Examples	1413
2.31.6 Advantages of Tree-Based Cluster Ensembles	1416
2.31.7 Conclusion	1417
References	1417
Index to Volume 2	1421
e9780444527028v3	1451
Series page	1453
Title page	1454
Copyright page	1455
Contents of Volume 3	1456
Contributors to Volume 3	1458
Preface	1462
Editors in Chief	1463
Contents of All Volumes	1464
Section Editors	1468
3.01 Calibration Methodologies	1472
3.01.1 Introduction	1472
3.01.2 Univariate Regression	1474
3.01.3 Multivariate Calibration	1476
3.01.4 The Eigenvector (Singular Vector) Basis Set	1480
3.01.5 PCR	1481
3.01.6 Generic Multivariate Calibration in the Eigenvector Basis Set	1484
3.01.7 Selecting Meta-Parameters via the Bias/Variance (Harmony) and Parsimony Balance	1484
3.01.8 Tikhonov Regularization and RR	1487
3.01.9 PLS	1493
3.01.10 PCR, RR, and PLS Intermodel Comparisons	1494
3.01.11 Other Multivariate Calibration Methods	1496
Appendix 1: Regression Coefficients	1498
Appendix 2: MATLAB SVD function	1498
Appendix 3: PCR	1498
Appendix 4: RR	1499
Appendix 5: PLS	1499
References	1500
3.02 Regression Diagnostics	1504
Symbols	1505
3.02.1 Introduction	1506
3.02.2 Formulation of MLR	1508
3.02.3 Diagnostics for MLR	1519
3.02.4 Useful Plots in Regression Diagnostics	1539
3.02.5 MATLAB Implementation of Regression Diagnostics	1554
Appendix 1 The OLS Solution Using Mean-Centered Data	1555
Appendix 2 The Confidence Region of OLS Coefficients and Experimental Design	1555
Appendix 3 Singular-Value Decomposition	1556
References	1557
3.03 Validation and Error	1562
Symbols	1562
3.03.1 Introduction	1563
3.03.2 Terminology	1564
3.03.3 Calibrations of Increasing Order	1566
3.03.4 Uncertainty Estimation	1571
3.03.5 Figures of Merit	1575
3.03.6 Validation	1580
Description of Data Sets and MATLAB Scripts	1583
Acknowledgment	1584
References	1584
3.04 Preprocessing Methods	1592
Symbols	1593
3.04.1 Introduction	1594
3.04.2 Methods for Calibration Data Set Selection	1597
3.04.3 Signal Correction Methods	1609
3.04.4 Methods for Dimensionality Reduction	1657
3.04.5 General Conclusion	1698
Acknowledgments	1698
References	1698
3.05 Variable Selection	1704
Symbols	1704
Glossary	1705
3.05.1 Introduction	1706
3.05.2 Criteria for Evaluating Variable Subsets without the Explicit Construction of a Model	1707
3.05.3 Criteria for Evaluating Variable Subsets on the Basis of the Regression Results	1713
3.05.4 Variable Selection Techniques	1719
3.05.5 Case Studies	1745
3.05.6 Limitations of Variable Selection	1751
References	1752
3.06 Missing Data	1756
Symbols	1756
3.06.1 Introduction	1758
3.06.2 Statistical Methods for Handling Missing Data	1760
3.06.3 Multivariate Calibration Model Building with Missing Data	1770
3.06.4 Software	1783
References	1783
3.07 Robust Calibration	1786
Symbols	1786
3.07.1 Introduction	1787
3.07.2 Location and Covariance Estimation	1788
3.07.3 Linear Calibration in Low Dimensions	1790
3.07.4 Principal Component Analysis	1794
3.07.5 Linear Calibration in High Dimensions	1799
3.07.6 Classification	1803
3.07.7 Multiway Analysis	1807
3.07.8 Software and Data Availability	1810
References	1810
3.08 Transfer of Multivariate Calibration Models	1816
Symbols	1816
3.08.1 Introduction	1817
3.08.2 How a Multivariate Calibration Model Can Become Invalid	1818
3.08.3 Strategies that Can Be Used before the Model Is Implemented	1820
3.08.4 Instrument Standardization Methods	1822
3.08.5 Selection of the Standardization Samples	1837
3.08.6 Calibration Transfer without Standardization	1838
3.08.7 Conclusions	1843
3.08.8 Software	1845
Acknowledgment	1845
References	1845
3.09 Three-Way Calibration	1850
3.09.1 Introduction	1851
3.09.2 The Trilinear Model	1854
3.09.3 Historical Algorithms of Trilinear Calibration	1856
3.09.4 Alternating Least Squares Methods	1860
3.09.5 Data Pretreatment and Preliminary Analyses	1864
3.09.6 Data Validation and Postprocessing	1866
3.09.7 Figures of Merit	1868
3.09.8 Example Analyses	1869
Appendix 1	1877
Appendix 2	1880
Appendix 3	1881
References	1881
3.10 Model-Based Data Fitting	1884
Symbols	1884
3.10.1 Introduction	1885
3.10.2 Data	1885
3.10.3 Models and Parameters	1887
3.10.4 Chemical Models	1889
3.10.5 Fitting, Sum of Squares, Parameters	1893
3.10.6 Other Issues	1902
References	1905
3.11 Kernel Methods	1908
Symbol	1908
3.11.1 Introduction	1908
3.11.2 Linear and Nonlinear Support Vector Machine Classifiers	1909
3.11.3 Support Vector Machine Regression	1913
3.11.4 Wider Use of the Kernel Trick	1915
3.11.5 Function Estimation in Reproducing Kernel Hilbert Spaces	1916
3.11.6 Least Squares Support Vector Machines: Regression and Classification	1916
3.11.7 Further Extensions: Kernel PCA/CCA/PLS	1918
Acknowledgments	1920
References	1920
3.12 Linear Approaches for Nonlinear Modeling	1924
Symbols	1924
3.12.1 Introduction	1924
3.12.2 Locally Weighted Regression	1924
3.12.3 Radial Basis Function Neural Networks	1926
3.12.4 Radial Basis Function Partial Least Squares Regression	1929
3.12.5 Neural Fuzzy Systems	1930
3.12.6 Conclusions	1931
References	1932
3.13 Other Methods in Nonlinear Regression	1934
Symbols	1934
3.13.1 Introduction	1934
3.13.2 Classification and Regression Trees	1936
3.13.3 Multivariate Adaptive Regression Splines	1938
3.13.4 Projection Pursuit Regression	1942
3.13.5 Illustrative Example	1943
References	1945
3.14 Neural Networks	1948
Symbols	1948
3.14.1 Introduction	1949
3.14.2 A Brief History	1950
3.14.3 The Artificial Neuron as the Basic Computational Unit	1951
3.14.4 From the Neuron to the Network	1953
3.14.5 Conclusions	1974
References	1974
3.15 Classification: Basic Concepts	1978
3.15.1 Introduction	1978
3.15.2 Linear Discriminant Functions	1979
3.15.3 Higher Levels of Classification	1981
3.15.4 Conclusion	1984
References	1985
3.16 Statistical Discriminant Analysis	1988
Symbols	1988
3.16.1 Introduction	1988
3.16.2 Canonical Discriminant Analysis	1990
3.16.3 Linear Discriminant Analysis	1995
3.16.4 Quadratic Discrimination	1996
3.16.5 Shrinkage and Covariance Stabilization	1997
3.16.6 Classification in High Dimensions	2006
3.16.7 Summary	2008
References	2009
3.17 Decision Tree Modeling in Classification	2012
3.17.1 Introduction	2013
3.17.2 Decision Tree Induction	2015
3.17.3 Examples	2022
3.17.4 Extending Decision Trees	2029
3.17.5 Generalizing Feature Space Partitioning	2031
3.17.6 Alternative Extensions and Uses of Decision Tree Modeling	2035
3.17.7 Software	2037
References	2038
3.18 Feed-Forward Neural Networks	2042
Symbols	2042
3.18.1 Introduction	2042
3.18.2 The Perceptron	2043
3.18.3 Multilayer Perceptrons	2044
3.18.4 Optimization of Weights	2046
3.18.5 Practical Considerations	2049
3.18.6 Applications of Feed-Forward Neural Networks	2052
3.18.7 Conclusion	2055
References	2056
3.19 Validation of Classifiers	2058
Symbols	2058
3.19.1 Introduction	2058
3.19.2 Assessing Predictive Ability Using a Validation Set	2061
3.19.3 Assessing Predictive Ability Using Cross-Validation and Bootstrapping	2061
3.19.4 Confounding	2062
3.19.5 Practical Considerations	2068
References	2069
3.20 Feature Selection: Introduction	2072
Symbols	2072
3.20.1 Introduction	2072
3.20.2 Data-Driven Science	2073
3.20.3 SIMCA	2074
3.20.4 Wavelets	2075
3.20.5 Genetic Algorithms	2075
3.20.6 Conclusions	2076
References	2076
3.21 Multivariate Approaches: UVE-PLS	2080
Symbols	2080
3.21.1 Introduction	2080
3.21.2 Uninformative Variable Elimination by PLS Algorithm	2081
3.21.3 Examples	2083
3.21.4 Other Applications	2087
References	2088
3.22 Multivariate Approaches to Classification using Genetic Algorithms	2090
Symbols	2090
3.22.1 Introduction	2090
3.22.2 Genetic Algorithms	2093
3.22.3 PCKaNN	2095
3.22.4 Ordinal Classes	2099
3.22.5 Robustification of PCKaNN	2099
3.22.6 Incorporation of Transverse Learning in PCKaNN	2100
3.22.7 Applications of the Pattern Recognition GA	2101
3.22.8 Conclusion	2114
References	2115
3.23 Feature Selection in the Wavelet Domain: Adaptive Wavelets	2118
Symbols	2118
3.23.1 Introduction	2120
3.23.2 Wavelets	2121
3.23.3 Statistical Methods Utilizing Adaptive Wavelets	2130
3.23.4 Applications of Adaptive Wavelets	2137
3.23.5 Concluding Remarks	2147
References	2148
3.24 Robust Multivariate Methods in Chemometrics	2152
Symbols	2153
3.24.1 Introduction	2154
3.24.2 Designing Robust Multivariate Estimators	2158
3.24.3 Robust Regression	2161
3.24.4 Robust Alternatives to Principal Component Analysis	2171
3.24.5 Robust Alternatives to Partial Least Squares	2173
3.24.6 Robust Approaches to Discriminant Analysis	2181
3.24.7 Validation	2186
References	2190
Index to Volume 3	2194
e9780444527028v4	2226
Series page	2228
Title page	2229
Copyright page	2230
Contents of Volume 4	2231
Contributors to Volume 4	2233
Preface	2237
Editors in Chief	2238
Contents of All Volumes	2239
Section Editors	2243
4.01 Representative Sampling, Data Quality, Validation – A Necessary Trinity in Chemometrics	2247
Symbols	2247
4.01.1 Introduction: Sampling of Heterogeneous Lots	2248
4.01.2 Heterogeneity	2248
4.01.3 Types of Sampling Errors Versus Practical Sampling	2250
4.01.4 Representative Mass Reduction	2254
4.01.5 Process Sampling (1-D Lots)	2255
4.01.6 Sampling Errors – Summary	2261
4.01.7 Seven Sampling Unit Operations	2262
4.01.8 Data Quality	2262
4.01.9 Validation in Chemometrics	2263
References	2264
4.02 Multivariate Statistical Process Control and Process Control, Using Latent Variables	2267
4.02.1 Introduction	2268
4.02.2 Traditional SPC Charts	2269
4.02.3 Latent Variable Based Process Monitoring	2275
4.02.4 Handling Future Observations with Missing Data	2286
4.02.5 Adaptive Latent Variable Models	2287
4.02.6 Batch Process Monitoring	2288
4.02.7 Monitoring Transitions in Continuous Processes	2292
4.02.8 Multistage Operations – Mutiblock Analysis	2293
4.02.9 Process Control Using Latent Variable Methods	2294
4.02.10 MIA for MSPC and Control	2295
4.02.11 Concluding Remarks	2296
References	2297
4.03 Environmental Chemometrics	2301
Symbols	2301
4.03.1 Introduction	2301
4.03.2 Pattern Recognition	2302
4.03.3 Mixture Resolution Problem	2304
4.03.4 Regression	2308
4.03.5 Multivariate Calibration Methods	2309
4.03.6 Factor Analysis	2310
4.03.7 Probability Estimates	2312
4.03.8 Summary	2316
References	2316
4.04 Application of Chemometrics to Food Chemistry	2321
4.04.1 Introduction	2321
4.04.2 History	2323
4.04.3 Overview	2324
4.04.4 Objectives and Possible Improvements	2348
Data Sets	2362
References	2370
4.05 Chemometrics in QSAR	2375
4.05.1 Introduction	2375
4.05.2 Short History of QSAR and Molecular Descriptors	2377
4.05.3 Chemometrics and QSAR Modeling	2379
4.05.4 Specific QSAR Approaches	2384
4.05.5 Molecular Descriptors	2390
4.05.6 Molecular Descriptor Selection	2397
4.05.7 Principles for QSAR Modeling	2403
4.05.8 Conclusions	2410
References	2410
4.06 Spectroscopic Imaging	2419
Symbols	2419
4.06.1 Introduction	2419
4.06.2 Introduction to NIR Imaging	2420
4.06.3 Chemometrics and NIR Imaging	2421
4.06.4 Example Applications	2428
4.06.5 Conclusion	2436
References	2436
4.07 Spectral Map Analysis of Microarray Data	2443
4.07.1 Introduction	2443
4.07.2 Microarray Technology	2444
4.07.3 Data Analysis of Microarrays	2445
4.07.4 Factor Analytic Thinking	2446
4.07.5 The Distinction between Size and Shape	2449
4.07.6 The Biplot Graphic	2450
4.07.7 Logarithms	2450
4.07.8 Singular Value Decomposition	2451
4.07.9 The Golub Data Set	2452
4.07.10 Principal Component Analysis	2452
4.07.11 Correspondence Analysis	2454
4.07.12 Spectral Map Analysis	2455
4.07.13 Interpretation of Spectral Maps Using Interactive Graphics	2457
4.07.14 The Size–Contrast Diagram	2459
4.07.15 Conclusion	2460
Acknowledgments	2460
References	2460
4.08 Analysis of Megavariate Data in Functional Genomics	2467
Symbols	2468
4.08.1 Introduction	2468
4.08.2 Molecular Basis of Functional Genomics	2469
4.08.3 Important Considerations in Functional Genomics	2473
4.08.4 Data Analysis	2479
4.08.5 Concluding Remarks	2516
Acknowledgment	2517
Appendix	2518
A.1 Matlab Codes for FDR Adjustment of Significance Test in PLSR	2518
A.2 Matlab Codes for Efficient Regression Algorithm for Megavariate Data	2519
A.3 Matlab Codes for Bootstrap of GEMANOVA Model with One Component	2520
References	2521
4.09 Systems Biology	2525
Symbols	2525
4.09.1 Introduction	2526
4.09.2 Study Setup	2531
4.09.3 Data Preprocessing	2536
4.09.4 Data Analysis	2538
4.09.5 Metabolite Identification	2545
4.09.6 Interpretation and Visualization	2548
References	2552
4.10 Chemometrics Role within the PAT Context: Examples from Primary Pharmaceutical Manufacturing	2559
Symbols	2559
4.10.1 Introduction	2559
4.10.2 NIRS and Chemometrics in API Production	2565
4.10.3 Case Studies	2570
4.10.4 Conclusions	2595
References	2598
4.11 Smart Sensors	2603
4.11.1 Introduction	2603
4.11.2 Toward ‘Smart Sensors’ Featuring On-Sensor Chemometrics	2603
4.11.3 Toward Robust Chemometrics for Spectroscopic Analyses	2604
4.11.4 Toward Spectrochemical Imaging	2616
4.11.5 Future Perspectives of Chemometrics for ‘Smart Sensors’	2617
References	2620
4.12 Chemometric Analysis of Sensory Data	2623
Symbols	2624
4.12.1 Introduction	2624
4.12.2 The Methodology of Sensory Analysis	2627
4.12.3 Experimental Design	2633
4.12.4 Data Formats and Pretreatment	2636
4.12.5 Graphical Displays of Data	2637
4.12.6 Univariate Statistics	2641
4.12.7 Multivariate Analysis	2643
4.12.8 Other Methodology in Sensory Analysis	2659
4.12.9 Case Studies	2661
4.12.10 Discussion	2664
References	2665
4.13 Chemometrics in Electrochemistry	2671
4.13.1 Introduction	2671
4.13.2 Experimental Design and Optimization	2672
4.13.3 Data Preparation and Transformation	2676
4.13.4 Data Exploration and Sample Classification	2678
4.13.5 Determination of Concentrations and Calibration	2679
4.13.6 Knowledge-Based Expert Systems	2697
4.13.7 Conclusions	2699
Acknowledgments	2700
References	2701
4.14 Chemoinformatics	2705
Symbols	2706
4.14.1 Introduction	2706
4.14.2 The Origins and Scope of Chemoinformatics	2706
4.14.3 Teaching Computers Chemistry: Data Input Problems	2709
4.14.4 In Silico Chemistry: Data Processing and Data Output Problems	2717
4.14.5 Internet Resources for Chemistry and Chemoinformatics	2744
4.14.6 Conclusions and Further Trends	2745
4.14.7 Sources of Further Information and Advice	2746
References	2746
4.15 High-Performance GRID Computing in Chemoinformatics	2753
Glossary	2754
4.15.1 Introduction	2754
4.15.2 Grid Components and Concepts	2758
4.15.3 Interfacing with the Grid	2767
4.15.4 Grid Security and Integrity	2770
4.15.5 Migrating Existing Applications for the Grid	2773
4.15.6 Grid Computing in the Laboratory	2777
4.15.7 Case Study: Grid-Based Benchmarking of QSAR Data	2779
References	2782
Cumulative Index	2787

Comprehensive Chemometrics

Additional Information

Book Details

Abstract

Table of Contents

Contact Us

Quick Navigation