Generalized memory-type estimators for time-based surveys: simulation experience and empirical results with birth weight dataset

Using auxiliary information at an estimation stage has an important role in forming estimators with better precision. This precision can further be increased using prior information available in any other forms. In this paper, we utilized a hybrid exponentially weighted moving average statistic to form new generalized ratio-and product-type estimators for a population mean in simple random sampling without replacement. This statistics utilizes information from current surveys and past surveys in the form of a hybrid exponentially weighted moving average. We derived expressions of mean square errors and biases for the suggested estimators. Further, we obtained mathematical conditions under which the suggested estimators will perform better than the existing estimators. We supported our results through a simulation and an empirical study.