ESTIMATION OF MEAN IN PRESENCE OF MISSING DATA UNDER TWO-PHASE SAMPLING SCHEME

To estimate the population mean with imputation i.e. the technique of substitutingmissing data, there are a number of techniques available in literature like Ratio method ofimputation, Compromised method of imputation, Mean method of imputation, Ahmed method ofimputation, F-T method of imputation, and so on. If population mean of auxiliary information isunknown then these methods are not useful and the two-phase sampling is used to obtain thepopulation mean. This paper presents some imputation methods of for missing values in twophasesampling. Two different sampling designs in two-phase sampling are compared underimputed data. The bias and m.s.e of suggested estimators are derived in the form of populationparameters using the concept of large sample approximation. Numerical study is performed overtwo populations using the expressions of bias and m.s.e and efficiency compared with Ahmedestimators.