Robust HEWMA-type estimators for population mean under non-normality

In the field of data science, the performance of a least square estimator is highly affected if the variable under study departs from normality or has the presence of outliers. This study proposes robust HEWMA-type estimators for the population mean by considering some auxiliary information under a non-normality assumption for a time-based survey. The proposed estimators are found to have the minimum mean square errors and biases compared to other relevant estimators for the long-tailed symmetric family. We have investigated the robustness properties of the proposed estimator in presence of the outliers. Finally, the authors provide a simulation study and a real-life application to support the theoretical outcomes of the proposed estimators over the relevant estimators.