Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

On the consistency of supervised learning with missing values

Abstract : In many application settings, the data have missing entries which make analysis challenging. An abundant literature addresses missing values in an inferential framework: estimating parameters and their variance from incomplete tables. Here, we consider supervised-learning settings: predicting a target when missing values appear in both training and testing data. We show the consistency of two approaches in prediction. A striking result is that the widely-used method of imputing with a constant, such as the mean prior to learning is consistent when missing values are not informative. This contrasts with inferential settings where mean imputation is pointed at for distorting the distribution of the data. That such a simple approach can be consistent is important in practice. We also show that a predictor suited for complete observations can predict optimally on incomplete data, through multiple imputation. Finally, to compare imputation with learning directly with a model that accounts for missing values, we analyze further decision trees. These can naturally tackle empirical risk minimization with missing values, due to their ability to handle the half-discrete nature of incomplete variables. After comparing theoretically and empirically different missing values strategies in trees, we recommend using the "missing incorporated in attribute" method as it can handle both non-informative and informative missing values.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-02024202
Contributor : Erwan Scornet Connect in order to contact the contributor
Submitted on : Monday, June 29, 2020 - 7:04:06 PM
Last modification on : Saturday, June 26, 2021 - 3:42:19 AM

Files

Identifiers

  • HAL Id : hal-02024202, version 3
  • ARXIV : 1902.06931

Citation

Julie Josse, Nicolas Prost, Erwan Scornet, Gaël Varoquaux. On the consistency of supervised learning with missing values. 2020. ⟨hal-02024202v3⟩

Share

Metrics

Record views

170

Files downloads

1358