Abstract
Modern statistical analysis often encounters massive datasets with ultrahigh-dimensional features. In this work, we develop a subsampling approach for feature screening with massive datasets. The approach is implemented by repeated subsampling of massive data and can be used for analyzing tasks with memory constraints. To conduct the procedure, we first calculate an R-squared screening measure (and related sample moments) based on subsamples. Second, we consider three methods to combine the local statistics. In addition to the simple average method, we design a jackknife debiased screening measure and an aggregated moment screening measure. Both approaches reduce the bias of the subsampling screening measure and therefore increase the accuracy of the feature screening. Last, we consider a novel sequential sampling method, that is more computationally efficient than the traditional random sampling method. The theoretical properties of the three screening measures under both sampling schemes are rigorously discussed. Finally, we illustrate the usefulness of the proposed method with an airline dataset containing 32.7 million records.
Supplementary Materials
Supplementary_Material.pdf: This document provides the extensions of the proposed method, the proofs of the theoretical results in the main text, and some additional simulation results. Appendix A reports some extensions and discussions of the proposed method. Appendix B contains the detailed proofs of the theoretical results of the AVS measure. Appendix C contains the detailed proofs of the main theorems and Lemmas developed in section 3.1-3.3 of the main text. In particular, it contains the proofs of theorems 1, 2, 3, 4, and 5 and Lemmas 1 and 2 of the main text. Appendix D contains the detailed proofs of screening consistency developed in sections 3.4 of the main text. In particular, it contains the proofs of theorems 5 and 6 and Lemma 3 of the main text. Appendix E provides technical lemmas which are useful to prove the results in section 3 of the main text. Finally, Appendix F contains some additional numerical results.
Code.zip: This file is the python code for the proposed method. Please see the “README.md” in the file for using the code.