57
33

Goodness-of-fit Tests for high-dimensional Gaussian linear models

Abstract

Let (Y,(Xi)iI)(Y,(X_i)_{i\in\mathcal{I}}) be a zero mean Gaussian vector and VV be a subset of I\mathcal{I}. Suppose we are given nn i.i.d. replications of the vector (Y,X)(Y,X). We propose a new test for testing that YY is independent of (Xi)iI\V(X_i)_{i\in \mathcal{I}\backslash V} conditionally to (Xi)iV(X_i)_{i\in V} against the general alternative that it is not. This procedure does not depend on any prior information on the covariance of XX or the variance of YY and applies in a high-dimensional setting. It straightforwardly extends to test the neighbourhood of a Gaussian graphical model. The procedure is based on a model of Gaussian regression with random Gaussian covariates. We give non asymptotic properties of the test and we prove that it is rate optimal (up to a possible log(n)\log(n) factor) over various classes of alternatives under some additional assumptions. Besides, it allows us to derive non asymptotic minimax rates of testing in this setting. Finally, we carry out a simulation study in order to evaluate the performance of our procedure.

View on arXiv
Comments on this paper