SAS for Data Analysis: Intermediate Statistical Methods (Statistics and Computing)

This book is intended for use as the textbook in a second course in applied statistics that covers topics in multiple regression and analysis of variance at an intermediate level. Generally, students enrolled in such courses are p- marily graduate majors or advanced undergraduate students from a variety of disciplines. These students typically have taken an introductory-level s- tistical methods course that requires the use a software system such as SAS for performing statistical analysis. Thus students are expected to have an - derstanding of basic concepts of statistical inference such as estimation and hypothesis testing. Understandably, adequate time is not available in a ?rst course in stat- tical methods to cover the use of a software system adequately in the amount of time available for instruction. The aim of this book is to teach how to use the SAS system for data analysis. The SAS language is introduced at a level of sophistication not found in most introductory SAS books. Important features such as SAS data step programming, pointers, and line-hold spe- ?ers are described in detail. The powerful graphics support available in SAS is emphasized throughout, and many worked SAS program examples contain graphic components. Preface 6 Contents 8 Introduction to the SAS Language 12 Introduction 12 Basic Language: Rules and Syntax 16 Creating SAS Data Sets 19 The INPUT Statement 22 SAS Data Step Programming Statements and Their Uses 27 Data Step Processing 35 More on INPUT Statement 42 Use of pointer controls 42 The trailing @ line-hold specifier 44 The trailing @@ line-hold specifier 46 Use of RETAIN statement 47 The use of line pointer controls 49 Using SAS Procedures 51 Exercises 59 More on SAS Programming and Some Applications 66 More on the DATA and PROC Steps 66 Reading data from files 67 Combining SAS data sets 69 Saving and retrieving permanent SAS data sets 75 User-defined informats and formats 80 Creating SAS data sets in procedure steps 85 SAS Procedures for Computing Statistics 90 The UNIVARIATE procedure 92 The FREQ procedure 99 Some Useful Base SAS Procedures 114 The PLOT procedure 115 The CHART procedure 124 The TABULATE procedure 130 Exercises 133 Statistical Graphics Using SAS/GRAPH 139 Introduction 139 An Introduction to SAS/GRAPH 139 Useful SAS/GRAPH procedures 140 GPLOT procedure 140 GCHART procedure 143 Writing SAS/GRAPH programs 146 Quantile Plots 156 Empirical Quantile-Quantile Plots 161 Theoretical Quantile-Quantile Plots or Probability Plots 164 Profile Plots of Means or Interaction Plots 169 Two-Dimensional Scatter Plots and Scatter Plot Matrices 173 Two-Dimensional Scatter Plots 173 Scatter Plot Matrices 176 Histograms, Bar Charts, and Pie Charts 179 Other SAS Procedures for High-Resolution Graphics 185 Exercises 191 Statistical Analysis of Regression Models 197 An Introduction to Simple Linear Regression 197 Simple linear regression using PROC REG 199 Lack of fit test using PROC ANOVA 205 Diagnostic use of case statistics 207 Prediction of new y values using regression 214 An Introduction to Multiple Regression Analysis 218 Multiple regression analysis using PROC REG 221 Case statistics and residual analysis 227 Residual plots 232 Examining relationships among regression variables 235 Types of Sums of Squares Computed in PROC REG and PROC GLM 241 Model comparison technique and extra sum of squares 241 Types of sums of squares in SAS 243 Subset Selection Methods in Multiple Regression 245 Subset selection using PROC REG 251 Other options available in PROC REG for model selection 259 Inclusion of Squared Terms and Product Terms in Regression Models 261 Including interaction terms in the model 262 Comparing slopes of regression lines using interaction 263 Analysis of models with higher-order terms with PROC REG 264 Exercises 271 Analysis of Variance Models 284 Introduction 284 Treatment Structure 287 Experimental Designs 288 Linear Models 289 One-Way Classification 291 Using PROC ANOVA to analyze one-wayclassifications 300 Making preplanned (or a priori) comparisons using PROC GLM 306 Testing orthogonal polynomials using contrasts 311 One-Way Analysis of Covariance 318 Using PROC GLM to perform one-way covariance analysis 321 One-way covariance analysis: Testing for equal slopes 330 A Two-Way Factorial in a Completely Randomized Design 337 Analysis of a two-way factorial using PROC GLM 340 Residual analysis and transformations 345 Two-Way Factorial: Analysis of Interaction 347 Two-Way Factorial: Unequal Sample Sizes 355 Two-Way Classification: Randomized Complete Block Design 367 Using PROC GLM to analyze a RCBD 370 Using PROC GLM to test for nonadditivity 376 Exercises 378 Analysis of Variance: Random and Mixed Effects Models 397 Introduction 397 One-Way Random Effects Model 401 Using PROC GLM to analyze one-way random effects models 404 Using PROC MIXED to analyze one-way random effects models 408 Two-Way Crossed Random Effects Model 415 Using PROC GLM and PROC MIXED to analyze two-way crossed random effects models 418 Randomized complete block design: Blocking when treatment factors are random 425 Two-Way Nested Random Effects Model 426 Using PROC GLM to analyze two-way nested random effects models 429 Using PROC MIXED to analyze two-way nested random effects models 433 Two-Way Mixed Effects Model 435 Two-way mixed effects model: Randomized complete blocks design 438 Two-way mixed effects model: Crossed classification 449 Two-way mixed effects model: Nested classification 461 Models with Random and Nested Effects for More Complex Experiments 473 Models for nested factorials 474 Models for split-plot experiments 480 Analysis of split-plot experiments using PROC GLM 482 Analysis of split-plot experiments usingPROC MIXED 489 Exercises 496 APPENDICES 510 SAS/GRAPH 510 Introduction 510 SAS/GRAPH Statements 520 Goptions statement 523 SAS/GRAPH global statements 523 Printing and Exporting Graphics Output 533 Tables 535 References 554 Index 557 Statistical Learning from a Regression Perspective considers statistical learning applications when interest centers on the conditional distribution of the response variable, given a set of predictors, and when it is important to characterize how the predictors are related to the response. As a first approximation, this is can be seen as an extension of nonparametric regression. Among the statistical learning procedures examined are bagging, random forests, boosting, and support vector machines. Response variables may be quantitative or categorical. Real applications are emphasized, especially those with practical implications. One important theme is the need to explicitly take into account asymmetric costs in the fitting process. For example, in some situations false positives may be far less costly than false negatives. Another important theme is to not automatically cede modeling decisions to a fitting algorithm. In many settings, subject-matter knowledge should trump formal fitting criteria. Yet another important theme is to appreciate the limitation of one's data and not apply statistical learning procedures that require more than the data can provide. The material is written for graduate students in the social and life sciences and for researchers who want to apply statistical learning procedures to scientific and policy problems. Intuitive explanations and visual representations are prominent. All of the analyses included are done in R. Richard Berk is Distinguished Professor of Statistics Emeritus from the Department of Statistics at UCLA and currently a Professor at the University of Pennsylvania in the Department of Statistics and in the Department of Criminology. He is an elected fellow of the American Statistical Association and the American Association for the Advancement of Science and has served in a professional capacity with a number of organizations such as the Committee on Applied and Theoretical Statistics for the National Research Council and the Board of Directors of the Social Science Research Council. His research has ranged across a variety of applications in the social and natural sciences Remarkable advances in computation and data storage and the ready availability of huge data sets have been the keys to the growth of the new disciplines of data mining and machine learning, while the enormous success of the Human Genome Project has opened up the field of bioinformatics. These exciting developments, which led to the introduction of many innovative statistical tools for high-dimensional data analysis, are described here in detail. The author takes a broad perspective; for the first time in a book on multivariate analysis, nonlinear methods are discussed in detail as well as linear methods. Techniques covered range from traditional multivariate methods, such as multiple regression, principal components, canonical variates, linear discriminant analysis, factor analysis, clustering, multidimensional scaling, and correspondence analysis, to the newer methods of density estimation, projection pursuit, neural networks, multivariate reduced-rank regression, nonlinear manifold learning, bagging, boosting, random forests, independent component analysis, support vector machines, and classification and regression trees. Another unique feature of this book is the discussion of database management systems. This book is appropriate for advanced undergraduate students, graduate students, and researchers in statistics, computer science, artificial intelligence, psychology, cognitive sciences, business, medicine, bioinformatics, and engineering. Familiarity with multivariable calculus, linear algebra, and probability and statistics is required. The book presents a carefully-integrated mixture of theory and applications, and of classical and modern multivariate statistical techniques, including Bayesian methods. There are over 60 interesting data sets used as examples in the book, over 200 exercises, and many color illustrations and photographs. "This book is an integrated treatment of applied statistical methods, presented at an intermediate level, and the SAS programming language. It serves as an advanced introduction to SAS as well as how to use SAS for the analysis of data arising from many different experimental and observational studies. While there are many introductory texts on SAS programming, statistical methods texts that solely make use of SAS as the software of choice for the analysis of data are rare. While this is understandable from a marketability point of view, clearly such texts will serve the need of many thousands of students and professionals who desire to learn how to use SAS beyond the basic introduction they usually receive from taking an introductory statistics course. More recently, several authors in statistical methodology have begun to incorporate SAS in their texts but these books are limited to more specialized subjects." "Many of the standard topics covered in statistical methods texts supplemented by advanced material more suited for a second course in applied statistics are included, so that specific aspects of SAS procedures can be illustrated. Brief but instructive reviews of the statistical methodologies used are provided, and then illustrated with analysis of data sets used in well-known statistical methods texts. Particular attention is devoted to discussions of models used in each analysis because the authors believe that it is important for users to have not only an understanding of how these models are represented in SAS but also because it helps in the interpretation of the SAS output produced."--Jacket

statistical Learning From A Regression Perspective Considers Statistical Learning Applications When Interest Centers On The Conditional Distribution Of The Response Variable, Given A Set Of Predictors, And When It Is Important To Characterize How The Predictors Are Related To The Response. As A First Approximation, This Is Can Be Seen As An Extension Of Nonparametric Regression. Among The Statistical Learning Procedures Examined Are Bagging, Random Forests, Boosting, And Support Vector Machines. Response Variables May Be Quantitative Or Categorical.

real Applications Are Emphasized, Especially Those With Practical Implications. One Important Theme Is The Need To Explicitly Take Into Account Asymmetric Costs In The Fitting Process. For Example, In Some Situations False Positives May Be Far Less Costly Than False Negatives. Another Important Theme Is To Not Automatically Cede Modeling Decisions To A Fitting Algorithm. In Many Settings, Subject-matter Knowledge Should Trump Formal Fitting Criteria. Yet Another Important Theme Is To Appreciate The Limitation Of One’s Data And Not Apply Statistical Learning Procedures That Require More Than The Data Can Provide.

the Material Is Written For Graduate Students In The Social And Life Sciences And For Researchers Who Want To Apply Statistical Learning Procedures To Scientific And Policy Problems. Intuitive Explanations And Visual Representations Are Prominent. All Of The Analyses Included Are Done In R.

This book is devoted to the mathematical foundation of boundary integral equations. The combination of?nite element analysis on the boundary with these equations has led to very e?cient computational tools, the boundary element methods (see e.g., the authors [139] and Schanz and Steinbach (eds.) [267]). Although we do not deal with the boundary element discretizations in this book, the material presented here gives the mathematical foundation of these methods. In order to avoid over generalization we have con?ned ourselves to the treatment of elliptic boundary value problems. The central idea of eliminating the?eld equations in the domain and - ducing boundary value problems to equivalent equations only on the bou- ary requires the knowledge of corresponding fundamental solutions, and this idea has a long history dating back to the work of Green [107] and Gauss [95, 96]. Today the resulting boundary integral equations still serve as a major tool for the analysis and construction of solutions to boundary value problems. "This book is devoted to the basic mathematical properties of solutions to boundary integral equations and presents a systematic approach to the variational methods for the boundary integral equations arising in elasticity, fluid mechanics, and acoustic scattering theory. It may also serve as the mathematical foundation of the boundary element methods. The latter have recently become extremely popular and efficient computational tools in applications. The authors are well known for their fundamental work on boundary integral equations and related topics, This book is a major scholarly contribution to the modern theory of boundary integral equations and should be accessible and useful to a large community of mathematical analysts, applied mathematicians, engineers and scientists."--Jacket. Funded By Dsu Title Iii 2007-2012. Introduction And Preview -- Data And Databases -- Random Vectors And Matrices -- Nonparametric Density Estimation -- Model Assessment And Selection In Multiple Regression -- Multivariate Regression -- Linear Dimensionality Reduction -- Linear Discriminant Analysis -- Recursive Partitioning And Tree-based Methods -- Artificial Neural Networks -- Support Vector Machines -- Cluster Analysis -- Multidimensional Scaling And Distance Geometry -- Committee Machines -- Latent Variable Models For Blind Source Separation -- Non-linear Dimensionality Reduction And Manifold Learning -- Correspondence Analysis. By Alan Julian Izenman. Includes Bibliographical References (pages 667-707) And Indexes. This is the first book on multivariate analysis to look at large data sets which describes the state of the art in analyzing such data. Material such as database management systems is included that has never appeared in statistics books before.