Asymptotic properties of the balanced repeated replication method for sample quantiles (Q1206725)

This paper deals with the estimation of variance of sample quantiles from complex survey data. The main purpose is to prove the asymptotic consistency of a general class of variance estimators based on a balanced selection of subsamples, which includes as special cases e.g. the method of \textit{M. Gurney} and \textit{R. S. Jewett} [J. Am. Stat. Assoc. 70, 819- 821 (1975; Zbl 0322.62014). The consistency results also hold when balanced subsampling is replaced by random subsampling. As a key technical prerequisite, a Bahadur-type representation for sample quantiles in stratified random sampling is proved.