EcoSta 2021: Start Registration
View Submission - EcoSta2021
A0324
Title: High-dimensional classifiers under the strongly spiked eigenvalue model Authors:  Aki Ishii - Tokyo University of Science (Japan) [presenting]
Kazuyoshi Yata - University of Tsukuba (Japan)
Makoto Aoshima - University of Tsukuba (Japan)
Abstract: One of the features of modern data is that the data dimension is extremely high. However, the sample size is relatively small. We call such data HDLSS data. In HDLSS situations, new theories and methodologies are required to develop for statistical inferences. We note that eigenvalues of high-dimensional data grow very rapidly depending on the dimension. There are two types of high-dimensional eigenvalue models: the strongly spiked eigenvalue (SSE) and non-SSE (NSSE) models. We consider high-dimensional classification under the SSE model. We give new classifiers by using a data transformation technique. We show that our classifiers have preferable properties in theory. Finally, we check the performances of our classifiers in simulations.