Efficient Imputation Methods for Estimating Population Proportion in Diagonal Systematic Sampling

Accurate estimation of proportions is essential for effective decision-making, resource management, and ensuring the reliability of statistical analyses. It underpins many critical processes across various domains, making it a foundational aspect of data analysis and interpretation. In this paper, we have proposed a regression-type and exponential type imputation methods which are free of unknown parameters for estimating missing values or non-responses while estimating population proportion under diagonal systematic sampling design. Imputation methods are crucial in data analysis and machine learning because they address the issue of missing data. Some key significance includes improved data quality, preservation of sample size, bias reduction and enhanced predictive power. The estimators of the proposed imputation methods were derived. The properties (biases and MSEs) of the class of estimators of the proposed imputation methods were derived up to first order approximation. Results of numerical illustration using simulated data revealed that the proposed estimators are more efficient and practicable than exiting estimators considered in the study.