Kumar Anoop, Emam Walid, Tashkandy Yusra
Department of Statistics, Central University of Haryana, Mahendergarh, Haryana, 123031, India.
Department of Statistics, Amity University, Lucknow, 226028, India.
Heliyon. 2024 Aug 14;10(16):e36090. doi: 10.1016/j.heliyon.2024.e36090. eCollection 2024 Aug 30.
With an emphasis on memory-type approaches, this study presents a class of estimators specifically designed for estimating population variation in simple random sampling (SRS). The term 'memory-type' pertaining to the use of exponentially weighted moving averages (EWMA) statistic for the estimation, which utilizes the current and past information in temporal surveys. The study provides expressions for the bias and mean square error (MSE) of these estimators and establishes conditions under which their efficiency represses the conventional and other memory-type estimators. The theoretical findings are reinforced through a comprehensive simulation study conducted on hypothetically sampled populations. Additionally, the effectiveness of the proposed estimators is demonstrated utilizing real-life population data. The findings of simulation and real data application show the superiority of the proposed memory type estimator over the existing usual and memory type estimators.
本研究着重于记忆型方法，提出了一类专门为估计简单随机抽样（SRS）中的总体方差而设计的估计量。“记忆型”一词涉及使用指数加权移动平均（EWMA）统计量进行估计，该统计量在时间调查中利用了当前和过去的信息。该研究给出了这些估计量的偏差和均方误差（MSE）的表达式，并确定了它们的效率优于传统估计量和其他记忆型估计量的条件。通过对假设抽样总体进行的全面模拟研究，强化了理论研究结果。此外，利用实际总体数据证明了所提出估计量的有效性。模拟和实际数据应用的结果表明，所提出的记忆型估计量优于现有的常规估计量和记忆型估计量。
