Multiple imputation through variational autoencoders

Missing values are ubiquitous in clinical and social science data. Incomplete data not only leads to loss of information but can also introduce bias, which poses a significant challenge for data analysis. Various imputation procedures were designed to handle incomplete data under different missingness mechanisms. Rubin (1977) introduced multiple imputation to attain valid inference from data with ignorable nonresponse. Some techniques and R packages are developed to implement multiple imputations, such as MICE, Amelia and MissForest. However, the running time of imputation using these methods can be excessive for large datasets. We propose a scalable multiple imputation method based on variational and denoising autoencoders. Our R package mivae is built using the tensorflow package in R, which enables fast computation and thus provides a scalable solution for missing data. In this presentation, I will demonstrate some features of the R package mivae and compare the performance of several commonly used multiple imputation techniques. Multiple imputation inference will also be discussed.

Nov 26, 2019 1:00 PM — Nov 28, 2019 3:00 PM