The Ascend Of Imputation Information Models - A Review of The Models and Reporting

Abstract

Missing data are unavoidable in real-world settings and has drawn significant attention in the statistical literature. It can lead to potentially biased results of the research study if handled inappropriately.   The objective of this study is to offer better perspective on missing data mechanism, different approaches of data imputation techniques, and metrics to assess performance of the widely used data imputation methods. This review also provides insights into the reasons of missing data and its effect on loss of information in epidemiological and clinical research. Further, it may also assist data scientists in the choice of suitable technique of data imputation while executing a data mining task.