Improved Mixed Estimator Using Two Auxiliary Variables For Full Extreme Maximum And Minimum Values In Single Phase Sampling

The use of multiple auxiliary variables has been established to improve precision in the estimators of ratio, regression and product respectively. However, the presence of extreme values in the distribution could annul such efficiency Olatayo et al. (2020). Extreme values could be small or minimum, large or maximum values. This study had developed a ratio-cum-regression estimator with two auxiliary variables, correlation coefficient and coefficient of variation under two types of extreme values in the distribution. This study considers full extreme value cases which assumed that both the study and two auxiliary variables had extreme values present in their distributions. Theoretical, empirical and percentage relative efficiency analyses were carried out for Full High and Maximum Extreme Values (FHMaEV) and Full Low and Minimum Extreme Values cases (FLMiEV). The analysis showed that the developed estimator is efficient over the reviewed estimators.