Does imputation matter? Benchmark for predictive models
This work addresses a practical issue for data scientists and ML practitioners by providing a comparative analysis to guide imputation method selection, though it is incremental as it builds on existing imputation techniques.
The paper tackles the problem of evaluating the impact of different data imputation methods on predictive model performance by conducting a systematic empirical benchmark on real-life classification tasks, finding that the choice of imputation method can significantly affect model accuracy, with variations of up to 10% across datasets.
Incomplete data are common in practical applications. Most predictive machine learning models do not handle missing values so they require some preprocessing. Although many algorithms are used for data imputation, we do not understand the impact of the different methods on the predictive models' performance. This paper is first that systematically evaluates the empirical effectiveness of data imputation algorithms for predictive models. The main contributions are (1) the recommendation of a general method for empirical benchmarking based on real-life classification tasks and the (2) comparative analysis of different imputation methods for a collection of data sets and a collection of ML algorithms.