CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Model-free variable selection and screening with matrix-valued predictors Authors:  Yuexiao Dong - Temple University (United States) [presenting]
Zeda Li - Baruch College (United States)
Abstract: A novel framework is introduced for model-free variable selection with matrix-valued predictors. To test the importance of rows, columns, and submatrices of the predictor matrix in terms of predicting the response, three types of hypotheses are formulated under a unified framework. In the fixed-dimensional setting, an asymptotic test as well as a permutation test are proposed to approximate the distribution of the test statistics under the null hypotheses. In the high-dimensional setting, the proposed test statistics can be used for marginal screening. The effectiveness of the proposed methods are evaluated through extensive numerical studies and an application to the electroencephalography (EEG) dataset.