Topics in optimal design (Q5952483)

This book is a follow-up to \textit{K.R. Shah} and \textit{B.K. Sinha}, Theory of optimal designs. Lect. Notes Stat. 54 (1989; Zbl 0688.62043). Unlike the previous book, this one covers a wide range of topics in both discrete and continuous optimal designs. (From the cover). In fact, the present volume constitutes a collection of recent research results by the four authors, spread over a variety of quite different topics. A survey of the contents is given in the introductory chapter 1, Scope of the monograph. Chapters 2 and 3 deal with optimal designs for regression models on both symmetric and asymmetric design regions. This covers also designs for models with random coefficients. Special emphasis is laid on the fact that optimal designs can be chosen with minimal support according to what the authors call the `de la Garza phenomenon'. Combinatorial aspects are investigasted in chapter 4 for optimal designs in analysis of variance models with additional continuous covariates: optimal designs for covariates' models with structured intercept parameter. In chapter 5, the comparatively new DS-optimality (stochastic distance optimality) criterion is introduced which measures the probabilities of deviations. Relations to other more established criteria are exhibited. Trend resistant designs are constructed in chapter 6, and chapter 7 covers some additional selected topics, like optimal designs for competing effects models, split block designs, nested designs, incomplete three-way balanced designs, and optimal designs under heteroscedastic errors. As this is a research exposition, some complicated proofs are omitted while references are given to the corresponding original papers by the same authors. Each chapter is self-contained and can be read by its own. The whole volume is augmented by both a common author and subject index which help to find out the connections between the single topics.