KLATASDS – MOE, School of Statistics, East China Normal University, Shanghai, People's Republic of China
ffang@sfs.ecnu.edu.cn
KLATASDS – MOE, School of Statistics, East China Normal University, Shanghai, People's Republic of China
Modern scientific research and applications very often encounter ‘fragmentary data’ which brings big challenges to imputation and prediction. By leveraging the structure of response patterns, we propose a unified and flexible framework based on Generative Adversarial Nets (GAN) to deal with fragmentary data imputation and label prediction at the same time. Unlike most of the other generative model based imputation methods that either have no theoretical guarantee or only consider Missing Completed At Random (MCAR), the proposed FragmGAN has theoretical guarantees for imputation with data Missing At Random (MAR) while no hint mechanism is needed. FragmGAN trains a predictor with the generator and discriminator simultaneously. This linkage mechanism shows significant advantages for predictive performances in extensive experiments.
To cite this article: Fang Fang & Shenliao Bao (2024) FragmGAN: generative adversarial nets for fragmentary data imputation and prediction, Statistical Theory and Related Fields, 8:1, 15-28, DOI: 10.1080/24754269.2023.2272554
To link to this article: https://doi.org/10.1080/24754269.2023.2272554