Exponential logarithmic approach for estimation of population parameter in two-phase sampling
Author(s): Bhargob Gohain, Deepjan Gohain and BK Singh
Abstract: Imputation is a commonly used approach to substitute missing data in sample surveys to address the issue of lack of response. The objective of the current study is to develop various imputation techniques for mean estimation of study variable in situations where the population mean of the auxiliary variable is unknown. Three imputation strategies for the exponential-logarithmic type estimators along with their respective point estimators have been suggested. Also, some expressions for the bias and the mean square error of the proposed estimators have been derived. Furthermore, some numerical data sets have been used to test the percentage relative efficiency of the proposed estimators. A simulation study has also been done for the efficiency of the estimators. Higher percentage relative efficiency indicates better performance. The proposed exponential-logarithmic imputation techniques provide more efficient and accurate ways to handle non-response in surveys compared to traditional imputation methods.
Bhargob Gohain, Deepjan Gohain, BK Singh. Exponential logarithmic approach for estimation of population parameter in two-phase sampling. Int J Stat Appl Math 2025;10(6):134-144. DOI: 10.22271/maths.2025.v10.i6b.2067
