TWO STEPS CALIBRATION ESTIMATORS OF FINITE POPULATION MEAN IN TWO-STAGE STRATIFIED RANDOM SAMPLING WHEN AUXILIARY INFORMATION IS AVAILABLE AT PRIMARY STAGE UNIT LEVEL

In the present paper, an attempt has been made to develop two-steps calibration estimator of population mean in two-stage stratified random sampling, first by calibrating sampling design weight at psu level and second by calibrating stratum weight using known total of the auxiliary information at population level. A limited simulation study with real data has been conducted to examine the relative performance of the calibration estimators over the usual estimator of the population mean without using auxiliary information in two-stage stratified random sampling. Results of the simulation study show that two-steps calibration estimator has brought considerable improvement in the precision of the estimate of the population mean.