Exponential type estimator for estimating finite population mean with auxiliary variables under simple random sampling

Abstract

In this study, a ratio-product-cum-exponential type estimators for estimating the population mean in single phase sampling were proposed. The biases and Mean Square Errors (MSEs) of these estimators were obtained up to the first order of approximation. Theoretical and empirical comparative approach using real datasets and simulation study were investigated. The results showed that the  proposed estimators were more efficient than the sample mean, ratio, product, exponential ratio and product estimator. Furthermore,  the efficiency of the proposed estimators were investigated at different correlation levels and it was found that as the correlation  increases the efficiency also changes positively. This suggests that when the auxiliary and study variables are more strongly correlated,  the estimators become more efficient, reducing estimation errors and increasing precision.