In this paper, the authors describe the process and results of their evaluation of the performance of multiple popular statistical methods for imputing missing metric measurements of bioarcheological or forensic skeleton specimen analysis.
It is not uncommon for biological anthropologists to analyze incomplete bioarcheological or forensic skeleton specimens. As many quantitative multivariate analyses cannot handle incomplete data, missing data imputation or estimation is a common preprocessing practice for such data. Using William W. Howells' Craniometric Data Set and the Goldman Osteometric Data Set, the authors evaluated the performance of multiple popular statistical methods for inputting missing metric measurements. Results indicated that multiple imputation methods outperformed single imputation methods, such as Bayesian principal component analysis (BPCA). Multiple imputation with Bayesian linear regression implemented in the R package norm2, the Expectation–Maximization (EM) with Bootstrapping algorithm implemented in Amelia, and the Predictive Mean Matching (PMM) method and several of the derivative linear regression models implemented in mice, perform well regarding accuracy, robustness, and speed. Based on the findings of this study, the authors suggest a practical procedure for choosing appropriate imputation methods. Publisher Abstract Provided
Downloads
Similar Publications
- Examining Radicalization's Risk and Protective Factors: A Case-Control Study of Violent Extremists, Non-Violent Criminal Extremists, Non-offending Extremists & Regular Violent Offenders
- The Lethality Assessment Program 2.0: Adjusting Intimate Partner Violence Risk Assessment to Account for Strangulation Risk
- Assessment of Sexual Assault Kit (SAK) Evidence Selection Leading to Development of SAK Evidence Machine-Learning Model (SAK-ML Model)