Ranked set sampling. Theory and applications (Q1414905)

This is the first monograph on ranked set sampling (RSS). The basic premise for RSS is the assumption that a set of sampling units drawn from the population can be ranked by certain means rather cheaply without the actual measurement of the variable of interest \(X\). The simplest form of RSS is as follows. First, for a simple random sample (SRS) of size \(k\), the \(k\) sampling units are ranked w.r.t. \(X\). Then the unit with rank 1 is taken for the measurement of \(X\), and the remaining units of the sample are discarded. Next, another SRS of size \(k\) is drawn, the units of the ample are ranked by judgement, and the unit with rank 2 is taken for the measurement of \(X\) and the remaining units are discarded. This process is continued until a SRS of size \(k\) is taken and ranked, and the unit with rank \(k\) is taken for the measurement of \(X\). Thus one cycle is accomplished, and it repeats \(m\) times and yields a ranked set sample of size \(mk\). Statistical inferences based on RSS are investigated from the estimation of a population mean, inferences on smooth-functions-of-means, quantiles and density functions, distribution-free tests and regression analysis. Both parametric and nonparametric aspects of RSS are addressed. In most cases, the RSS estimates are more efficient than the corresponding SRS estimates. The book presents a happy mixture of theory and applications. Indeed, theoretical foundations are provided with mathematical rigor, and various methods are illustrated by real or simulated data. An essential part of the book contains original results obtained by the authors. The book will used as a common reference by statisticians as well as practitioners in agriculture, sociology, ecological and medical studies, etc. It can serve as a textbook for a short course at the graduate level. Contents: Chap. 1. Introduction; Chaprs. 2 and 3: Balanced ranked set sampling. I: Nonparametric, II: Parametric; Chap. 4: Unbalanced RSS and optimal designs; Chap. 5: Distribution-free tests with RSS; Chap. 6: RSS with concomitant variables; Chap. 7: RSS as data reduction tools; Chap. 8: Case studies