The process of developing predictive models includes many stages. Most resources focus on the modeling algorithms but neglect other critical aspects of the modeling process. This book describes techniques for finding the best representations of predictors for modeling and for nding the best subset of predictors for improving model performance. A variety of example data sets are used to illustrate the techniques along with R programs for reproducing the results.

Geocomputation with R is for people who want to analyze, visualize and model geographic data with open source software. It is based on R, a statistical programming language that has powerful data processing, visualization, and geospatial capabilities. The book equips you with the knowledge and skills to tackle a wide range of issues manifested in geographic data, including those with scientific, societal, and environmental implications. This book will interest people from many backgrounds, especially Geographic Information Systems (GIS) users interested in applying their domain-specific knowledge in a powerful open source language for data science, and R users interested in extending their skills to handle spatial data. The book is divided into three parts: (I) Foundations, aimed at getting you up-to-speed with geographic data in R, (II) extensions, which covers advanced techniques, and (III) applications to real-world problems. The chapters cover progressively more advanced topics, with early chapters providing strong foundations on which the later chapters build. Part I describes the nature of spatial datasets in R and methods for manipulating them. It also covers geographic data import/export and transforming coordinate reference systems. Part II represents methods that build on these foundations. It covers advanced map making (including web mapping), "bridges" to GIS, sharing reproducible code, and how to do cross-validation in the presence of spatial autocorrelation. Part III applies the knowledge gained to tackle real-world problems, including representing and modeling transport systems, finding optimal locations for stores or services, and ecological modeling. Exercises at the end of each chapter give you the skills needed to tackle a range of geospatial problems. Solutions for each chapter and supplementary materials providing extended examples are available at https://geocompr.github.io/geocompkg/articles/. Dr. Robin Lovelace is a University Academic Fellow at the University of Leeds, where he has taught R for geographic research over many years, with a focus on transport systems. Dr. Jakub Nowosad is an Assistant Professor in the Department of Geoinformation at the Adam Mickiewicz University in Poznan, where his focus is on the analysis of large datasets to understand environmental processes. Dr. Jannes Muenchow is a Postdoctoral Researcher in the GIScience Department at the University of Jena, where he develops and teaches a range of geographic methods, with a focus on ecological modeling, statistical geocomputing, and predictive mapping. All three are active developers and work on a number of R packages, including stplanr, sabre, and RQGIS.

This book offers a predominantly theoretical coverage of statistical prediction, with some potential applications discussed, when data and/ or parameters belong to a large or infinite dimensional space. It develops the theory of statistical prediction, non-parametric estimation by adaptive projection – with applications to tests of fit and prediction, and theory of linear processes in function spaces with applications to prediction of continuous time processes. This work is in the Wiley-Dunod Series co-published between Dunod (www.dunod.com) and John Wiley and Sons, Ltd.

Disk contains the library functions and documentation for use with Splus for Windows.

The changing focus and approach of geomorphic research suggests that the time is opportune for a summary of the state of discipline. The number of peer-reviewed papers published in geomorphic journals has grown steadily for more than two decades and, more importantly, the diversity of authors with respect to geographic location and disciplinary background (geography, geology, ecology, civil engineering, computer science, geographic information science, and others) has expanded dramatically. As more good minds are drawn to geomorphology, and the breadth of the peer-reviewed literature grows, an effective summary of contemporary geomorphic knowledge becomes increasingly difficult. The fourteen volumes of this Treatise on Geomorphology will provide an important reference for users from undergraduate students looking for term paper topics, to graduate students starting a literature review for their thesis work, and professionals seeking a concise summary of a particular topic. Information on the historical development of diverse topics within geomorphology provides context for ongoing research; discussion of research strategies, equipment, and field methods, laboratory experiments, and numerical simulations reflect the multiple approaches to understanding Earth’s surfaces; and summaries of outstanding research questions highlight future challenges and suggest productive new avenues for research. Our future ability to adapt to geomorphic changes in the critical zone very much hinges upon how well landform scientists comprehend the dynamics of Earth’s diverse surfaces. This Treatise on Geomorphology provides a useful synthesis of the state of the discipline, as well as highlighting productive research directions, that Educators and students/researchers will find useful. Geomorphology has advanced greatly in the last 10 years to become a very interdisciplinary field. Undergraduate students looking for term paper topics, to graduate students starting a literature review for their thesis work, and professionals seeking a concise summary of a particular topic will find the answers they need in this broad reference work which has been designed and written to accommodate their diverse backgrounds and levels of understanding Editor-in-Chief, Prof. J. F. Shroder of the University of Nebraska at Omaha, is past president of the QG&G section of the Geological Society of America and present Trustee of the GSA Foundation, while being well respected in the geomorphology research community and having won numerous awards in the field. A host of noted international geomorphologists have contributed state-of-the-art chapters to the work. Readers can be guaranteed that every chapter in this extensive work has been critically reviewed for consistency and accuracy by the World expert Volume Editors and by the Editor-in-Chief himself No other reference work exists in the area of Geomorphology that offers the breadth and depth of information contained in this 14-volume masterpiece. From the foundations and history of geomorphology through to geomorphological innovations and computer modelling, and the past and future states of landform science, no "stone" has been left unturned!

A practical and accessible introduction to the bootstrap method——newly revised and updated Over the past decade, the application of bootstrap methods to new areas of study has expanded, resulting in theoretical and applied advances across various fields. Bootstrap Methods, Second Edition is a highly approachable guide to the multidisciplinary, real-world uses of bootstrapping and is ideal for readers who have a professional interest in its methods, but are without an advanced background in mathematics. Updated to reflect current techniques and the most up-to-date work on the topic, the Second Edition features: The addition of a second, extended bibliography devoted solely to publications from 1999–2007, which is a valuable collection of references on the latest research in the field A discussion of the new areas of applicability for bootstrap methods, including use in the pharmaceutical industry for estimating individual and population bioequivalence in clinical trials A revised chapter on when and why bootstrap fails and remedies for overcoming these drawbacks Added coverage on regression, censored data applications, P-value adjustment, ratio estimators, and missing data New examples and illustrations as well as extensive historical notes at the end of each chapter With a strong focus on application, detailed explanations of methodology, and complete coverage of modern developments in the field, Bootstrap Methods, Second Edition is an indispensable reference for applied statisticians, engineers, scientists, clinicians, and other practitioners who regularly use statistical methods in research. It is also suitable as a supplementary text for courses in statistics and resampling methods at the upper-undergraduate and graduate levels.

In this dissertation, we present several methodologies for mining spatial and time-sequence data obtained in diverse domains. We first propose a new spatial randomness test and classification method for binary spatial data with specific application to the detection and identification of spatial defect patterns on semiconductor wafer maps. We present the generalized join-count (JC)-based statistic as an alternative approach, and derive a procedure to determine the optimal weights of JC-based statistics. In the proposed methodology, a spatial correlogram, which transforms binary spatial data into time-sequence data, is used as a novel feature to detect spatial autocorrelation and classify spatial defect patterns on the wafer maps. Secondly, we propose a novel distance measure, denoted weighted dynamic time warping (WDTW), for time series classification and clustering problems. The dynamic time warping (DTW) algorithm has been extensively used as a distance measure in combination with the distance-based classifiers. However, the DTW algorithm ignores the relative importance of the phase distance between points in a time series, possibly leading to misclassification. Therefore, we propose a WDTW distance measure which does account for the relative importance of each point in terms of the phase distance between the time series points. Thirdly, we propose a wavelet-based anomaly detection procedure to detect any possible process fault with time-sequence data that have some local variations even under normal working conditions. To handle the large number of parameters in both the mean and variance models, we have developed the wavelet-based mean and variance thresholding procedure to extract a few important wavelet coefficients that may explain local variations in the time domain. Finally, we propose a kernel-based regression with lagged dependent variables. Kernel-based regression techniques are extensively used for exploring the nonlinearity of data in a relatively easy procedure involving the use of various kernel functions. However, the major drawback of current kernel-based regression techniques is their underlying assumption that there is no autocorrelation in the residuals of observations. To avoid this problem, we propose a kernel-based regression model with lagged dependent variables (LDVs), considering autocorrelations of both the response variables and the nonlinearity of data.

Although standard mixed effects models are useful in a range of studies, other approaches must often be used in correlation with them when studying complex or incomplete data. Mixed Effects Models for Complex Data discusses commonly used mixed effects models and presents appropriate approaches to address dropouts, missing data, measurement errors, censoring, and outliers. For each class of mixed effects model, the author reviews the corresponding class of regression model for cross-sectional data. An overview of general models and methods, along with motivating examples After presenting real data examples and outlining general approaches to the analysis of longitudinal/clustered data and incomplete data, the book introduces linear mixed effects (LME) models, generalized linear mixed models (GLMMs), nonlinear mixed effects (NLME) models, and semiparametric and nonparametric mixed effects models. It also includes general approaches for the analysis of complex data with missing values, measurement errors, censoring, and outliers. Self-contained coverage of specific topics Subsequent chapters delve more deeply into missing data problems, covariate measurement errors, and censored responses in mixed effects models. Focusing on incomplete data, the book also covers survival and frailty models, joint models of survival and longitudinal data, robust methods for mixed effects models, marginal generalized estimating equation (GEE) models for longitudinal or clustered data, and Bayesian methods for mixed effects models. Background material In the appendix, the author provides background information, such as likelihood theory, the Gibbs sampler, rejection and importance sampling methods, numerical integration methods, optimization methods, bootstrap, and matrix algebra. Failure to properly address missing data, measurement errors, and other issues in statistical analyses can lead to severely biased or misleading results. This book explores the biases that arise when naïve methods are used and shows which approaches should be used to achieve accurate results in longitudinal data analysis.