Contemporary Statistical Models for the Plant and Soil Sciences [With CD-ROM]

ISBN-13: 9781584881117 / Angielski / Twarda / 2001 / 760 str.

PH.D. Oliver Schabenberger; Francis J. Pierce

Opis

Integrating modern statistics with practical problems in soil and crop science, Contemporary Statistical Models for the Plant and Soil Science provides real-life examples that illustrates the material. Advances in technology have significantly altered the type, amount, quality, timing, format, and the source of data available. This book shows the reader how to apply these techniques to their work. The techniques presented include: mixed linear models, nonlinear regression, and spatial statistics. A CD-ROM is included that allows the reader to work the problems on their own.

Recenzje

"This text [presents] many of the newer statistical modeling techniques for data analysis using examples familiar to plant and soil scientists…keeping the mathematical complexity to a minimum. I applaud the authors for their efforts to bring the current state of the area of statistical modeling into the realm of the plant and soil sciences."
--Clarence E. Watson, Experimental Statistics and Plant and Soil Sciences, Mississippi State University, USA

"My overall impression is that it is a superbly crafted text replete with many carefully chosen examples that instructively demonstrate contemporary models and modelling practices. The authors' attention to fine detail in the presentation of materials is evident in every chapter. Researchers, instructors, and students alike doubtlessly will find the snippets of SAS code and specially tailored macros to be of immense value when fitting data to the contemporary models described in this treatise."
--Timothy Gregoire, School of Forestry and Environmental Studies, Yale University, New Haven , USA

Spis treści

Statistical ModelsMathematical and Statistical ModelsFunctional Aspects of ModelsThe Inferential Steps ó Estimation and Testingt-Tests in Terms of Statistical ModelsEmbedding HypothesesHypothesis and Significance Testing ó Interpretation of the p-ValueClasses of Statistical ModelsData StructuresIntroductionClassification by Response TypeClassification by Study TypeClustered DataAutocorrelated DataFrom Independent to Spatial Data ó A Progression of ClusteringLinear Algebra ToolsIntroductionMatrices and VectorsBasic Matrix OperationsMatrix Inversion ó Regular and Generalized InverseMean, Variance, and Covariance of Random VectorsThe Trace and Expectation of Quadratic FormsThe Multivariate Gaussian DistributionMatrix and Vector DifferentiationUsing Matrix Algebra to Specify ModelsThe Classical Linear Model: Least Squares and AlternativesIntroductionLeast Squares Estimation and Partitioning of VariationFactorial ClassificationDiagnosing Regression ModelsDiagnosing Classification ModelsRobust EstimationNonparametric Regression Nonlinear ModelsIntroductionModels as Laws or ToolsLinear Polynomials Approximate Nonlinear ModelsFitting a Nonlinear Model to DataHypothesis Tests and Confidence IntervalsTransformationsParameterization of Nonlinear ModelsApplicationsGeneralized Linear ModelsIntroductionComponents of a Generalized Linear ModelGrouped and Ungrouped DataParameter Estimation and InferenceModeling an Ordinal ResponseOverdispersionApplicationsLinear Mixed Models for Clustered DataIntroductionThe Laird-Ware ModelChoosing the Inference SpaceEstimation and InferenceCorrelations in Mixed ModelsApplicationsNonlinear Models for Clustered DataIntroductionNonlinear and Generalized Linear Mixed ModelsTowards an Approximate Objective FunctionApplicationsStatistical Models for Spatial DataChanging the MindsetSemivariogram Analysis and EstimationThe Spatial ModelSpatial Prediction and the Kriging ParadigmSpatial Regression and Classification ModelsAutoregressive Models for Lattice DataAnalyzing Mapped Spatial Point PatternsApplicationsBibliography

O autorze

Francis J. Pierce, Oliver Schabenberger