Optimal Sparse Linear Prediction for Block-missing Multi-modality Data Without Imputation

• Published in: Journal of the American Statistical Association Vol. 115; no. 531; pp. 1406 - 1419
• Main Authors: Yu, Guan, Li, Quefeng, Shen, Dinggang, Liu, Yufeng
• Format: Journal Article
• Language: English
• Published: Alexandria Taylor & Francis 02.07.2020; Taylor & Francis Ltd
• Subjects: Alzheimer disease; Alzheimer's disease; Americans; Block-missing; Continuity (mathematics); Covariance matrix; dropouts; Dropping out; Huber's M-estimate; Imputing; Lasso; Linear prediction; Mathematical analysis; Matrix algebra; Matrix methods; Medical imaging; methodology; Missing data; Multi-modality; Neuroimaging; Prediction; Regression analysis; Sparse regression; Statistical analysis; Statistical methods; Statistics; variance covariance matrix

In modern scientific research, data are often collected from multiple modalities. Since different modalities could provide complementary information, statistical prediction methods using multimodality data could deliver better prediction performance than using single modality data. However, one spec...