Robust calibration for estimating the population mean using stratified random sampling

Estimating the population mean is of prime concern in many studies, and calibrations are popular choices. A robust calibration estimator estimates the mean using the minimum covariance determinant (MCD) and the minimum volume ellipsoid (MVE) estimations under stratified random sampling. Efficiency comparisons have been made between the robust calibration estimator and classical calibration estimator. Simulations and empirical results show that the proposed robust calibration estimator has a lower mean square error than the calibration estimators. When the relative efficiency and computation times are considered together, it is seen that the proposed robust calibration estimators based on MCD estimates are more efficient.