Answers are grounded in this paper and cite their sources.
Existing composite estimators based on the population total and the ratio estimator assume precise data, although survey data are often uncertain and imprecise. This paper reformulates the composite estimator of the population total under generalized interval statistics, which extends classical and interval statistics by modelling the indeterminate part of each observation explicitly. We derive the expectation, variance, optimal weight and proportional reduction in variance of the proposed estimator, and we show that it reduces to the classical composite estimator when no indeterminacy is present. A simulation study based on bivariate normal data with interval-valued uncertainty examines the effects of the sample size and the correlation, and its results differ markedly from those of classical statistics. An application to monthly electricity line loss data from District Okara, Pakistan, shows that classical statistics and interval midpoints either lose information or introduce bias relative to the proposed method. Several performance indices, including loss of information, relative underestimation, interval width, overlap and convergence indices, are used to compare the proposed method with classical statistics, interval statistics and an existing simulation-based approach. The proposed estimator performs favorably in terms of informativeness, convergence, and computational complexity, and it is recommended for the composite estimation of population totals under uncertainty
No takes yet. Share an insight, caveat, or question.
Muhammad Aslam (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: