Developing calibration estimators for population mean using robust measures of dispersion under stratified random sampling

Abstract In this paper, two modified, design-based calibration ratio-type estimators are presented. The suggested were developed under stratified random sampling using information on an auxiliary variable in the form of robust statistical measures, including Gini’s mean difference, Downton’s method and probability weighted moments. properties (biases MSEs) proposed studied up to terms first-order approximation by means Taylor’s Series approximation. theoretical results supported a simulation study conducted four bivariate populations generated normal, chi-square, exponential gamma populations. indicate that scheme is more precise than any others considered paper.