Multivariate Methods of Representing Relations in R for Prioritization Purposes

This monograph is multivariate, multi-perspective and multipurpose. We intend to be innovatively integrative through statistical synthesis. Innovation requires capacity to operate in ways that are not ordinary, which means that conventional computations and generic graphics will not meet the needs of an adaptive approach. Flexible formulation and special schematics are essential elements that must be manageable and economical.

Supports the assertion that multiple views of data have a greater prospect of revealing prominent patterns than single views This book provides a portal into the realms of the open source data analysis system called R though exposition by example The authors formulate approaches that are operable in R that support preliminary and progressive prioritization

Autorentext
Dr. Wayne L. Myers earned M.F. and Ph.D. degrees in forest ecology and forest entomology at the University of Michigan. He began his professional career in Canada as a research forest entomologist and biometrician. He then joined the faculty of forestry at Michigan State University specializing in biometrics and remote sensing.

Klappentext

This monograph is a four-fold featuring of adaptive analysis.

· First is data distillation and comparative coupling whereby the results of one analysis are fed forward into another analysis without necessarily returning directly to the original data matrix, and analytical avenues usually seen as alternatives are pursued in parallel with results being carried forward together as complementary comparatives.

· Second is the flexibility and suitability of the R© statistical software system for engaging in such adaptive and conjunctive statistical strategies. The intention is to provide an extensive entry into the realms of R using exploration by example whereby a demonstrative dataset of manageably moderate size is carried comparatively though the sequence of sections.

· Third is a major mission to introduce innovative methodologies for preliminary and/or partial prioritization that arise from partial order theory. We formulate functions in R that provide for generation and visualization of partial orderings based on combinations of criteria. These methods support etiological exploration for explanations that underlie apparent concurrence or conflict among multiple indicators of suitability or severity.

Fourth is delving more deeply into some multivariate methods such as principal components using the matrix methods available in R. R makes highly compact calls available for several such multivariate methods, but sometimes discernment demands delving into details.

Inhalt
Motivation and Computation.- Part I: Synergistic Scalings, Contingent Clustering and Distance Domains.- Suites of Scalings.- Rotational Rescaling and Disposable Dimensions.- Comparative Clustering for Contingent Collectives.- Distance Domains, Skeletal Structures and Representative Ranks.- Part II: Precedence and Progressive Prioritization.- Ascribed Advantage, Subordination Schematic and ORDIT Ordering.- Precedence Plots, Coordinated Crite4ria and Rank Relations.- Case Comparisons and Precedence Pools.- Distal Data and Indicator Interactions.- Landscape Linkage for Prioritizing Proximate Patches.- Constellations of Criteria.- Severity Setting for Human Health.- Part III: Transformation Techniques and Virtual Variates.- Matrix Methods for Multiple Measures.- Segregating Sets Along Directions of Discrimination.- Index.