Statistical Computation for Environmental Sciences in R: Lab Manual for Models for Ecological Data (Lab Manual) [Paperback] Clark, James S Paperback - 2007

"Clark brings emerging statistical approaches alive by putting the ecology first. Writing from the perspective of a field ecologist who must confront complex data without suppressing important detail, Clark describes new methods that are well matched to the richness of real ecological data. At last we have a text that makes these tools accessible to ecologists."--Stephen R. Carpenter, University of Wisconsin, Madison

"Jim Clark has been able to pitch his message just right; one can see the ecological forest and the statistical, distributional, and computational trees at the same time. By reading this book, statisticians will gain an appreciation for the complexity of models in the ecological and environmental sciences, and ecologists will see the potential for hierarchical statistical modeling in their research arenas. Clark explains his material extremely well, but he is also rigorous in his statistical developments."--Noel Cressie, Ohio State University

"Clark's book is monumental--I don't think there is any other source that provides this range of sources and methods. He presents a huge amount of useful material, focusing on the development and application of Bayesian hierarchical models for the analysis of ecological and environmental models. It's hard to imagine finding such a collection of information--the results of extensive experience with recent ecological, environmental, and statistical literature--in one place. And I heartily agree with the author's philosophical stances on simplicity and complexity, statistical pragmatism, and the need for common sense."--Benjamin Bolker, University of Florida

"I strongly believe that this is potentially a landmark book in ecology. Its integration of modern statistical methods and ecological theory and data is fundamentally new. The book will train ecologists and other quantitative scientists in the 'new modeling techniques' that are becoming ever more prevalent in their field. In particular, the book describes how one should deal with complicated problems in which there is uncertainty in data, model, and parameters. James Clark does a wonderful job of integrating modern likelihood-based statistical methods as well as describing and demonstrating the advantages of the Bayesian approach."--Christopher K. Wikle, University of Missouri, Columbia