Optimal class of memory type imputation methods for time-based surveys using EWMA statistics.

ABSTRACT: Time-based surveys often experience missing data due to several reasons, like non-response or data collection limitations. Imputation methods play an essential role in incorporating these missing values to secure the accuracy and reliability of the survey outcomes. This manuscript proposes some optimal class of memory type imputation methods for imputing missing data in time-based surveys by utilizing exponentially weighted moving average (EWMA) statistics. The insights into the optimal conditions for incorporating our proposed methods are provided. A comprehensive examination of the proposed method utilizing simulated and real-life datasets is conducted. Comparative analyses against the existing imputation methods exhibit the superior performance of our methods, particularly in the scenar

SUBMITTER: Kumar A 

PROVIDER: S-EPMC11519609 | biostudies-literature | 2024 Oct

REPOSITORIES: biostudies-literature