Statistical Methods: The Geometric Approach

David J. Saville, Graham R. Wood (Autoren)

XV, 560 Seiten

1997 | 3., corr. Printing
Springer Berlin (Hersteller)
978-3-540-97517-5 (ISBN)

Infos

This text is a novel exposition of the traditional workhorses of statistics: analysis of variance and regression. It is a novel in that it presents this core material for the first time in a consistent visual way. The exposition is aimed at students of any discipline who need to learn this material. The approach is based on the geometric methods uses by the founder of modern-day statistics, Sir Ronald Fisher. The text uses geometry to link the statistical methods of the underlying mathematics, while real-life case studies link theory and practice. Each new situation is introduced by a detailed analysis of a case study, including assumption checking and presentation of results. Computing is simply carried out in Minitab. The book is well illustrated with diagrams and graphs, and includes class exercises and a wealth of practical exercises. Readers can expect to benefit from a deeper understanding of statistical methods and their applications to real problems.

Inhalt

Basic Ideas: Introduction: The Geometric Tool Kit. The Statistical Tool Kit. Tool Kits At Work.- Introduction to Analysis of Variance: Single Population Questions. Questions About Two Populations. Questions About Several Populations. Orthogonal Contrasts: Class Comparisons. Factorial Contrasts. Polynomial Contrasts. Pairwise Comparisons.- Introducing Blocking: Randomized Block Design. Latin Square Design. Split Plot Design.- Fundamentals of Regression: Simple Regression.- Polynomial Regression.- Analysis of Covariance.- General Summary. Appendices: A: Unequal Replications. Two Populations. B: Unequal Replications. Several Populations. C: Alternative Factorial Notation. D: Regression Through the Origin. E: Confidence Intervals. T: Statistical Tables. References. Index.

Produkt-Details

• Reihe/Serie: Springer Texts in Statistics
• Zusatzinfo: 143 figs.
• Sprache: deutsch
• Gewicht: 946 g
• Einbandart: gebunden
• Themenwelt: Mathematik / Informatik ► Mathematik ► Wahrscheinlichkeit / Kombinatorik
• Schlagworte: HC/Mathematik/Wahrscheinlichkeitstheorie, Stochastik, Mathematische Statistik • Statistik
• ISBN-10: 3-540-97517-9 / 3540975179
• ISBN-13: 978-3-540-97517-5 / 9783540975175
• Zustand: Neuware