Improved novel estimation for estimation of population distribution function using auxiliary information under stratified sampling strategy

Abstract

The distribution function (DF) is a determinant functional parameter in many academic disciplines that use probability tools for data analysis, such as economics and medicine. Among other functions, it helps to estimate quantiles and other parameters. Therefore, strategies for its efficient estimation are needed. In this article, we have suggested improved classes of estimators for the estimation of population DF using auxiliary information. These include the conventional unbiased estimator, the difference-type estimator, and other well-known estimators. We derive numerical expressions for the corresponding bias and mean squared error (MSE) from the first-order approximation, aiming to evaluate their effectivness. When compared to some existing competitors, the suggested estimators performed much better in terms of minimum MSE and higher percentage relative efficiency (PRE).