A framework for testing different imputation methods for tabular datasets
Tabea Kossen, Michelle Livne, Vince I. Madai, Ivana Galinović, Dietmar Frey, Jochen B. Fiebach
The paper develops a framework to test imputation methods on a clinical stroke dataset and finds that listwise deletion performs poorly, while the best method depends on the missing data mechanism.
bioRxiv · Sep 17, 2019 · PreprintNeuroscience