Multivariate Analysis, Design of Experiments, and Survey Sampling / Edition 1

by Subir Ghosh

Overview

"Describes recent developments and surveys important topics in the areas of multivariate analysis, design of experiments, and survey sampling. Features the work of nearly 50 international leaders."

Product Details

ISBN-13:	9780824700522
Publisher:	CRC Press
Publication date:	04/29/1999
Series:	Statistics: A Series of Textbooks and Monographs
Edition description:	New Edition
Pages:	698
Product dimensions:	8.50(w) x 11.00(h) x (d)

Table of Contents

Sampling designs and prediction methods for Gaussian spatial processes; design techniques for probabilistic sampling of items with variable monetary value; small area estimation -a Bayesian perspective; Bayes sampling designs for selection procedures; cluster coordinated composites of diverse datasets on several spatial scales for designing extensive environmental sample surveys - prospectus on promising protocols; corrected confidence sets for sequentially designed experiments, II -examples; resampling marked point processes; graphical Markov models in multivariate analysis; robust regression with censored and truncated data; multivariate calibration; some consequences of random effects in multivariate survival models; a unified methodology for constructing multivariate autoregressive models; statistical model evaluation and information criteria; multivariate rank tests; asymptotic expansions of the distribution of some test statistics for elliptical populations; a review of variance estimators with extensions to multivariate nonparametric regression models; on affine invariant sign and rank tests in one and two sample multivariate problems; correspondence and component analysis; growth curve models; dealing with uncertainties in queues and networks of queues; optimal Bayesian design for a logistic regression model - geometric and algebraic approaches; structure of weighing mattrices of small order and weight.

Editorial Reviews

The collection of 22 articles describes recent developments, and surveys topics in the three areas of statistics listed in the title. Parametric, nonparametric, frequentist, Bayesian and bootstrap methods, spatial processes, point processes, Markov models, ranking and selection, robust regression, calibration, model selection, survival analysis, and queuing and networks are among the topics discussed. More specifically, the compilation provides a risk prediction model for optimally selecting inspection samples from shipments containing items of highly variable monetary value, reviews progress on obtaining approximations to sampling distributions in sequentially designed experiments, considers the effects of elliptical populations for test statistics from multivariate normal theory, and supplies a nonstandard introduction to correspondence analysis and some of its generalizations. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews