Imputation Uncertainty in Interpretable Machine Learning Methods
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908722934054912 |
|---|---|
| author | Golchian, Pegah Wright, Marvin N. |
| author_facet | Golchian, Pegah Wright, Marvin N. |
| contents | In real data, missing values occur frequently, which affects the interpretation with interpretable machine learning (IML) methods. Recent work considers bias and shows that model explanations may differ between imputation methods, while ignoring additional imputation uncertainty and its influence on variance and confidence intervals. We therefore compare the effects of different imputation methods on the confidence interval coverage probabilities of the IML methods permutation feature importance, partial dependence plots and Shapley values. We show that single imputation leads to underestimation of variance and that, in most cases, only multiple imputation is close to nominal coverage. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_17689 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Imputation Uncertainty in Interpretable Machine Learning Methods Golchian, Pegah Wright, Marvin N. Machine Learning Methodology In real data, missing values occur frequently, which affects the interpretation with interpretable machine learning (IML) methods. Recent work considers bias and shows that model explanations may differ between imputation methods, while ignoring additional imputation uncertainty and its influence on variance and confidence intervals. We therefore compare the effects of different imputation methods on the confidence interval coverage probabilities of the IML methods permutation feature importance, partial dependence plots and Shapley values. We show that single imputation leads to underestimation of variance and that, in most cases, only multiple imputation is close to nominal coverage. |
| title | Imputation Uncertainty in Interpretable Machine Learning Methods |
| topic | Machine Learning Methodology |
| url | https://arxiv.org/abs/2512.17689 |