Evaluation of missing data imputation methods for human osteometric measurements
Bibliographic Data
| ID | 24314422 |
|---|---|
| Authors | Jinyong Pang (USF Genomics & College of Public Health University of South Florida Tampa Florida USA), Xiaoming Liu (0000-0001-7904-1056, USF Genomics & College of Public Health University of South Florida Tampa Florida USA) |
| Year | 2023 |
| Volume | 181 |
| Issue | 4 |
| Pages | 666-676 |
| Publication date | 2023-08-01 |
| Open Access | No |
| Type | ARTICLE |
| Venue | American Journal of Biological Anthropology (JOURNAL) |
| Journal identifiers | ISSN: 2692-7691 • E-ISSN: 2692-7691 |
| Publisher | Wiley (PUBLISHER • GB) |
| DOI | 10.1002/ajpa.24787 |
| PMID | 37259623 |
| OpenAlex | W4378953040 |
| Language | EN |
| Citations received | 2 |
| References cited | 32 |
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, we evaluated the performance of multiple popular statistical methods for imputing 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, we suggest a practical procedure for choosing appropriate imputation methods.
Amelia · Bayesian principal component analysis · Bioarchaeology · craniometry · Forensic anthropology · imputation method comparison · mice · missing data imputation · multiple imputation · predictive mean matching · skeletal remains · Forensic Anthropology and Bioarchaeology Studies · Statistical Methods and Applications · Statistical Methods and Bayesian Inference
Missing data in bioarchaeology II
Technical note
Missing data in bioarchaeology I
Amelia II
MissMDA
Imputation with the R Package VIM
R
MissForest—non-parametric missing value imputation for mixed-type data
A Review of Hot Deck Imputation for Survey Non‐response
Maximum Likelihood from Incomplete Data Via the EM Algorithm
Mice
Multiple Imputation for Nonresponse in Surveys
Statistical Matching Using File Concatenation With Adjusted Weights and Multiple Imputations
Human body mass estimation
The macromorphoscopic databank
Limb bone bilateral asymmetry
Missing Data Imputation Using Morphoscopic Traits and Their Performance in the Estimation of Ancestry
How Many Imputations Do You Need? A Two-stage Calculation Using a Quadratic Rule
| Citations per year | 1 |
|---|---|
| Citation span | 2024 - 2025 (2) |
| Highly cited | No |