Efficient population mean estimation of sensitive traits using robust quantile regression and scrambled response methodology

Using robust regression approach for mean estimation in survey sampling with a single supplementary information is a well-established practice when there are outliers in the data set. The most popular methods for sensitive studies are regression estimation methods that use standard regression coefficients. When it comes to survey researchers, mean estimation is an important issue. Many researchers introduced a class of mean estimation methods utilizing data on two additional variables of information under simple random sampling (SRS), with the help of several non-conventional location measures and typical OLS. In this article, we propose a novel method for computing the population mean using robust quantile regression-type estimators for scrambled response model (SRM) under SRS. According to the results, the suggested estimator performs better than other estimators in the literature.