The Role of the Computer in Statistics David Cox Nuffield College, Oxford OXIINF, U.K. A classification of statistical problems via their computational demands hinges on four components (I) the amount and complexity of the data, (il) the specificity of the objectives of the analysis, (iii) the broad aspects of the approach to analysis, (ill) the conceptual, mathematical and numerical analytic complexity of the methods. Computational requi rements may be limiting in (I) and (ill), either through the need for special programming effort, or because of the difficulties of initial data management or because of the load of detailed analysis. The implications of modern computational developments for statistical work can be illustrated in the context of the study of specific probabilistic models, the development of general statistical theory, the design of investigations and the analysis of empirical data. While simulation is usually likely to be the most sensible way of investigating specific complex stochastic models, computerized algebra has an obvious role in the more analyti cal work. It seems likely that statistics and applied probability have made insufficient use of developments in numerical analysis associated more with classical applied mathematics, in particular in the solution of large systems of ordinary and partial differential equations, integral equations and integra-differential equations and for the ¢raction of "useful" in formation from integral transforms. Increasing emphasis on models incorporating specific subject-matter considerations is one route to bridging the gap between statistical ana.
Auflage
Softcover reprint of the original 1st ed. 1992
Sprache
Verlagsort
Zielgruppe
Für Beruf und Forschung
Research
Illustrationen
18
18 s/w Abbildungen
XVI, 579 p. 18 illus.
Maße
Höhe: 244 mm
Breite: 170 mm
Dicke: 32 mm
Gewicht
ISBN-13
978-3-662-26813-1 (9783662268131)
DOI
10.1007/978-3-662-26811-7
Schweitzer Klassifikation
Prologue.- Science, Data, Statistics and Computing.- I. Statistical Modelling.- Graphical Regression.- LAPACK: A Linear Algebra Library for High-Performance Computers.- PARELLA: Measurement of Latent Attitude via the Statistical Modelling of Response Processes.- On Fitting Non-Stationary Markov Point Process Models on GLIM.- Measuring Departure from Gaussian Assumptions in Spatial Processes.- Monte Carlo Simulation on Software Mutation Test-Case Adequacy.- Exploratory Analysis of Ordinal Repeated Measures.- II. Multivariate Analysis.- Scaling Factors in Generalised Procrustes Analysis.- Correspondance Table between Two Sets, One Partitioned, and Associated Tables.- DOMOUSE: Detection of Multivariate Outliers Using S Environment.- Algorithmic Approaches for Fitting Bilinear Models.- Ordinal Principal Component Analysis.- Computational Procedures for Estimation and Hypothesis Testing in Analyzing Two-Way Tables with Interaction and No Replication.- Computational Aspects of the Nearest Neighbor Statistics.- III. Classification and Discrimination.- Discrimination through the Regularized Nearest Cluster Method.- Segmentation Trees: A New Help Building Expert Systems and Neural Networks.- Classification by Quantification of Ranking Pattern.- Tree-Growing for the Multivariate Model: The RECPAM Approach.- On the Level of Regularization in Regularized Discriminant Analysis.- Additive Spline Discriminant Analysis.- A New Dynamic Programing Algorithm for Cluster Analysis.- StatLog: An Evaluation of Machine Learning and Statistical Algorithms.- Discriminant Analysis of Plasma Fusion Data.- Presentation of the Method of Density Slices: An Answer to the Multitendencies Problems.- A Two-Stage Predictive Splitting Algorithm in Binary Segmentation.- A Comparative Study of Two Methods ofDiscrete Regularized Discriminant Analysis.- IV. Symbolic and Relational Data.- From Data to Knowledge: Probabilist Objects for a Symbolic Data Analysis.- Comparing Numerical and Symbolic Clustering: Application to Pseudoperiodic Light Curves of Stars.- Description of a Knowledge Base of Symbolic Objects by Histograms.- An Efficient Neural Network by a Classification Tree.- V. Graphical Models.- Graphical Models for Dependencies and Associations.- Model Search in Contingency Tables by CoCo.- Stochastic Simulation in Mixed Graphical Association Models.- Recognizing Submodels of a Loglinear Model.- HModel: An X Tool for Global Model Search.- Assessing the Power of Model Selection Procedures used when Graphical Modelling.- VI. Time Series Models.- Computing Missing Values in Time Series.- Forecasting Singapore's GDP Growth with a Small-Scale Model.- Construction of State Space Models for Time Series Exhibiting Exponential Growth.- The Technique of ?-Transformation in Econometric Data Series.- Econometric Modelling on Micro-Computers: The Impact of the New Technological Features.- Two New Robust Methods for Time Series.- Forecasting Using a Semiparametric Model.- Computing ARMA Models with MATLAB.- Analyzing Treatment Effects - The WAMASTEX Approach to Paired Sample Data.- Fortune: Improving Forecasts by Tuning the Forecasting Process.- A Comparative Study of Outlier Detection and Missing Value Estimation Methods Applied to Time Series Transport Data.- Linking Judgements to Forecasting Models.- Numerical Computation of Exact Distributions for First Order Stochastic Difference Equations.- VII. Nonlinear Regression.- Estimation of Radioimmunoassay Data Using Robust Nonlinear Regression Methods.- An Artificial Intelligence Approach for Modeling in Nonlinear RegressionParametric Models.- Accurate Multivariable Numerical Derivatives for Inference in Nonlinear Regression.- Advantages of the Approximative Interpretation - An Algorithm and Program for Finding Solutions of Large Non-Linear Problems.- Providing for the Analysis of Generalized Additive Models within a System already Capable of Generalized Linear and Nonlinear Regression.- A Note on Local Sensitivity of Regression Estimates.- Parallel Model Analysis with Factorial Parameter Structure.- VIII. Robustness and Smoothing Techniques.- Time-Efficient Algorithms for Two Highly Robust Estimators of Scale.- Universal Consistency of Partitioning-Estimates of a Regression Function for Randomly Missing Data.- The Use of Slices in the LMS and the Method of Density Slices: Foundation and Comparison.- On Some Statistical Properties of Bézier Curves.- Trade Regression.- A Review on Smoothing Methods for the Estimation of the Hazard Rate Based on Kernel Functions.- Departures from Assumptions in Two-Phase Regression.- An Analysis of the Least Median of Squares Regression Problem.- Nonparametric Regression Methods for Smoothing Curves.- An Inference Procedure for the Minimum Sum of Absolute Errors Regression.- An Operator Method for Backfitting with Smoothing Splines in Additive Models.- Sensitivity Analysis in Structural Equation Models.- Methods for Robust Non-Linear Regression.- IX. Industrial Applications: Pharmaceutics and Quality Control.- Statistical Thinking in Total Quality Management: Implications for Improvement of Quality - with Ideas for Software Development.- Statistical Computation in the Pharmaceutical Industry.- Stochastic Simulations of Population Variability in Pharmacokinetics.- Computational Aspects in Uncertainty Analyses of Physiologically-Based PharmacokineticModels.- X. Bayesian Statistics.- Simulated Annealing in Bayesian Decision Theory.- Generalized Linear Models and Jeffreys Priors: An Iterative Weighted Least-Squares Approach.- Approaches to Tobit Models via Gibbs Sampling.- Bayesian Predictive Screening: An Efficient Approach for the Multivariate Binary, Ordinal Responses.- Author Index.- Table of Contents of Volume Two.- Author Index of Volume Two.
