Multiple Imputation Using Deep Denoising Autoencoders
share
Missing data is a well-recognized problem impacting all domains. State-of-the-art framework to minimize missing data bias is multiple imputation, for which the choice of an imputation model remains nontrivial. We propose a multiple imputation model based on overcomplete deep denoising autoencoders, capable of handling different data types, missingness patterns, missingness proportions and distributions. Evaluation on real life datasets shows our proposed model outperforms the state-of-the-art methods under varying conditions and improves the end of the line analytics.
READ FULL TEXT
