CMStatistics 2022: Start Registration
View Submission - CMStatistics
B1238
Title: Some tests of hypotheses for high-dimensional linear models Authors:  Rauf Ahmad - Uppsala University (Sweden) [presenting]
Abstract: A statistic for testing the parameter vector in a general linear model is presented when the dimension of the parameter vector is large, i.e., when the number of columns of the design matrix may exceed the number of independent rows. The distribution of the proposed statistic, obtained under a few mild assumptions, depends on a simple function of the eigenvalues of the known, fixed design matrix. Simulations are used to show the accuracy of the proposed theory. Applications on real data are demonstrated. Some extensions of the test for other models are also considered.