Abstract

The problem of missing values in the sample is very common in most surveys. Through this manuscript, we suggest a generalized exponential chain dual-to-ratio type imputation method and the corresponding point estimator of the finite population mean in a double sampling scheme with two auxiliary variables in case of missing data in the sample. The Bias and Mean Square Error (MSE) of the proposed estimator are derived in terms of parameters. Theoretical studies are carried out to verify the supremacy of the proposed point estimator over other existing estimators considered in this study. An empirical study has been done to compare the percent relative efficiency of the proposed class of estimators with other existing estimators by using one natural population data and another artificially generated population data, generated by Shukla and Thakur (2008) A simulation study have been also carried out among the proposed and existing estimators for numerical calculation of bias, mean square error and percent relative efficiencies. The numerical calculations are found similar in both the Empirical and simulation studies. The novelty of the study is that the proposed imputation method is more efficient than the existing imputation methods considered in this study.

Author: Deepjan Gohain, Prabudh Gogoi and B. K. Singh

Received on: April, 2023

Accepted on: February, 2025