Tight conditions for consistency of variable selection in the context of high dimensionality

TL;DR

This paper establishes precise conditions under which variable selection is consistent in high-dimensional nonparametric regression, revealing different regimes depending on the growth of relevant variables and ambient dimension.

Contribution

It provides tight, nonparametric conditions for consistent variable selection in high dimensions, extending understanding beyond linear models without assuming parametric forms.

Findings

Consistent variable selection is possible if log d is small compared to n when the intrinsic dimension is fixed.

When the number of relevant variables grows, consistent selection requires s+loglog d to be small compared to log n.

The paper derives minimax separation rates for variable selection in high-dimensional nonparametric models.

Abstract

We address the issue of variable selection in the regression model with very high ambient dimension, that is, when the number of variables is very large. The main focus is on the situation where the number of relevant variables, called intrinsic dimension, is much smaller than the ambient dimension d. Without assuming any parametric form of the underlying regression function, we get tight conditions making it possible to consistently estimate the set of relevant variables. These conditions relate the intrinsic dimension to the ambient dimension and to the sample size. The procedure that is provably consistent under these tight conditions is based on comparing quadratic functionals of the empirical Fourier coefficients with appropriately chosen threshold values. The asymptotic analysis reveals the presence of two quite different re gimes. The first regime is when the intrinsic dimension is fixed. In this case the situation in nonparametric regression is the same as in linear regression, that is, consistent variable selection is possible if and only if log d is small compared to the sample size n. The picture is different in the second regime, that is, when the number of relevant variables denoted by s tends to infinity as $n\to\infty$. Then we prove that consistent variable selection in nonparametric set-up is possible only if s+loglog d is small compared to log n. We apply these results to derive minimax separation rates for the problem of variable