2015 248 pages
Series: Chapman & Hall/CRC Monographs on Statistics & Applied Probability
978-1-48-224279-9
21st June 2015
Most approaches to statistical inference refer to the evidence of something being true or false. But no theory exists that defines what this evidence is as an explicit quantity, which can lead to a lack of confidence in the conclusions. This book presents a theory of statistical inference based on a definition of statistical evidence via "relative belief." It offers a new framework for conducting statistical analyses in important practical problems. The book outlines the problems of statistical inference, discusses the theory in detail, and provides examples of its application.
Statistical Problems. Probability. Statistical Evidence and Relative Belief. Choosing and Checking the Ingredients. Appendices. Bibliography. Index.
2015 349 pages
Series: Chapman & Hall/CRC Monographs on Statistics & Applied Probability
978-1-49-871216-3
30th June 2015
Written by leading experts, this book discusses new methods for dealing with high-dimensional data. It summarizes the actively developing field of statistical learning with sparsity. Covering matrix decomposition, graphical models, compressed sensing, and more, it will be of interest to people analyzing data in many scientific disciplines.
Introduction. Basic Lasso Idea for Regression. Examples and Applications to Other Models. Generalized Penalties. Convex Optimization and Algorithms. Matrix Decompositions, Approximations and Completion. Multivariate Analysis and Sparsity. Graphical Models. Compressed sensing, image denoising, and Other EE Topics. Theoretical Results on Consistency.
2015 384 pages
Series: Chapman & Hall/CRC Texts in Statistical Science
978-1-48-223703-0
24th August 2015
This book explores the interface between environmental epidemiology (EE) and spatio-temporal modelling. It is based on a course taught at UBC to graduate students from epidemiology, mathematics, and statistics, and to public health practitioners. It provides necessary background material in EE and then develops the spatio-temporal theory, primarily Bayesian hierarchical modeling, necessary to study EE problems. It includes many detailed worked examples with R software and other computational tools, such as WinBUGS and INLA. Each chapter has exercises and lab projects to enable use for self-study or as a course text.
Basic Epidemiology. Regression modeling in epidemiology. Bayesian methods. Bayesian Computation. Temporal Processes. Spatial Lattice and Point Processes. Disease Mapping. Point Referenced Spatial Data. Spatial Regression. Spatio-temporal Modelling.
2015 600 pages
Series: Chapman & Hall/CRC Texts in Statistical Science
978-1-49-872268-1
14th October 2015
This book covers a number of topics that are important in current measure theory and practice. It emphasizes nonparametric methods which can really only be implemented with modern computing power on large and complex data sets. In addition, it includes a large number of problems with more difficult ones appearing with hints and partial solutions for the instructor.
Statistical Theory & Methods
Statistical Computing
2015 350 page
Series: Chapman & Hall/CRC Monographs on Statistics & Applied Probability
Hardback:
978-1-43-988648-9
14th October 2015
This book delves into the authorsf work toward deeper understanding of statistical inference in terms of reasoning with uncertainty and meaningfulness of probabilistic inferential output. Focusing on a valid, prior-free probabilistic inferential framework called inferential models, the authors explain how to first identify the underlying source of uncertainty as an integral part of statistical modeling and then make probabilistic inference by calculating the predictable quantity in a statistically accurate way.
Introduction. Inferential Models. Conditional Inferential Models. Marginal Inferential Models. Inferential Models and Likelihood-Based Inference. Bibliography.
2015 504 pages
Series: Chapman & Hall/CRC Texts in Statistical Science
978-1-49-872583-5
14th November 2015
The special nature of discrete variables and frequency data vis-a-vis statistical graphics is now more widely accepted, and many of these methods (e.g., mosaic displays, fourfold plots, diagnostic plots for generalized linear models) have become, if not main stream, then at least more widely used in research and teaching. This book provides an accessible introduction to the major methods of categorical data analysis for data exploration, statistical testing and statistical models. As opposed to more theoretical books, the goal here is to help the reader to translate theory into practical application, by providing skills and software tools for carrying out these methods.
Working with Categorical Data. Fitting and Graphing Discrete Distributions. Two-Way Contingency Tables. Mosaic Displays for n-Way Tables. Correspondence Analysis. Loglinear and Logit Models. Logistic Regression. Generalized Linear Models. Regression Models for Count Data. Repeated Measures and Longitudinal Data. Classification and Regression Trees. Other Material.
2015 300 pages
Series: Chapman & Hall/CRC Monographs on Statistics & Applied Probability
Hardback:
978-1-46-656165-6
14th December 2015
Absolute risk is the probability of developing a specific disease over a specified time interval in the presence of competing causes of mortality. Although absolute risk is arguably more relevant to clinical decision making than "pure" risk, the development of appropriate statistical methods for estimating and applying absolute risk has lagged behind those for pure risk. This book focuses on the development, evaluation, and application of models of absolute risk.
ntroduction. Definitions and Basic Concepts for Survival Data in a Cohort without Covariates. Developing Absolute Risk Models from Cohort Data with Covariates. Estimating Absolute Risk from Case-Cohort and Nested Case-Control Data. Estimating Absolute Risk from Population-Based Case-Control and Registry Data. Evaluation of Adequacy of Model. Comparing Two Models. Special Topic: Disease Prognosis. Special Topic: Family-Based Designs