GENERALIZED CHAIN RATIO REGRESSION AND CHAIN REGRESSION ESTIMATORS FOR THE POPULATION MEAN USING TWO AUXILIARY CHARACTERS IN THE PRESENCE OF NON-RESPONSE
Abstract
Generalized chain ratio regression and chain regression estimators for the population mean using two auxiliary characters have been proposed in the presence of non-response. The expressions for the mean square error (MSE) of these estimators have been obtained for fixed sample sizes ( n′ and n ) and also for fixed costC ≤ C0 .The optimum values of the first phase sample size n′ and the sub sample of size n have been obtained for the fixed cost 0 C ≤ C and also for the specified precision V = 0 V ′ .