Bootstrap confidence intervals for finite population quantiles in the presence of auxiliary information

Article type: Research Article

Authors: Rueda, M.^{; *} | Martínez-Miranda, M.D. | Arcos, A.

Affiliations: Department of Statistics and O.R., University of Granada, Campus Fuentenueva, (18071) Granada, Spain

Correspondence: [*] Corresponding author. E-mail: mrueda@ugr.es.

Abstract: This paper considers the construction of bootstrap confidence intervals for estimating finite population quantiles. The intervals involve estimators which make use of the auxiliary information provided by an available auxiliary variable. The proposed bootstrap confidence intervals are validated under theoretical (asymptotic) and empirical points of view, indeed simulations with natural populations are carried out in order to compare the behavior of the proposed intervals with the obtained from standard normal approximations.

Keywords: Indirect estimators, percentile bootstrap, sampling design

DOI: 10.3233/MAS-2006-1407

Journal: Model Assisted Statistics and Applications, vol. 1, no. 4, pp. 279-290, 2006