AN IMPROVED ROBUST REGRESSION TYPE MEAN ESTIMATOR USING RE-DESCENDING M-ESTIMATOR UNDER TWO-STAGE SAMPLING APPROACH

Abstract

When the data includes some outliers, estimating the population average using the ordinary least squares (OLS) approach is not effective. In this article, we develop a novel technique for computing the population average using regression robust-type estimator in two phase sampling. The developed technique uses a robust regression type estimator called re-descending M-estimation. The mean square error (MSE) equation is generated using a first-order approximation and examined with existing estimating techniques in order to evaluate the efficacy of the new approach. Additionally, the proposed estimator's percentage relative efficiency (PRE) is determined compared to other estimators. The effectiveness of the proposed method is demonstrated using real data sets. According to the results, the developed estimator performs efficient than other existing estimators in the literature.