Convergence rates of least squares regression estimators with heavy-tailed errors

TL;DR

This paper analyzes the convergence rates of least squares estimators in nonparametric regression with heavy-tailed errors, revealing conditions under which the estimator performs optimally or poorly.

Contribution

It establishes the convergence rate of the LSE under heavy-tailed errors with a new multiplier inequality and explores the impact of error moments and covariate independence.

Findings

LSE converges at a rate depending on the entropy exponent and error moments.

Optimal convergence rate matches Gaussian error case when errors have sufficient moments.

Dependence between errors and covariates can cause arbitrarily slow convergence.

Abstract

We study the performance of the Least Squares Estimator (LSE) in a general nonparametric regression model, when the errors are independent of the covariates but may only have a $p$-th moment ($p\geq 1$). In such a heavy-tailed regression setting, we show that if the model satisfies a standard `entropy condition' with exponent $\alpha \in (0,2)$, then the $L_2$ loss of the LSE converges at a rate \begin{align*} \mathcal{O}_{\mathbf{P}}\big(n^{-\frac{1}{2+\alpha}} \vee n^{-\frac{1}{2}+\frac{1}{2p}}\big). \end{align*} Such a rate cannot be improved under the entropy condition alone. This rate quantifies both some positive and negative aspects of the LSE in a heavy-tailed regression setting. On the positive side, as long as the errors have $p\geq 1+2/\alpha$ moments, the $L_2$ loss of the LSE converges at the same rate as if the errors are Gaussian. On the negative side, if $p<1+2/\alpha$, there are (many) hard models at any entropy level $\alpha$ for which the $L_2$ loss of the LSE converges at a strictly slower rate than other robust estimators. The validity of the above rate relies crucially on the independence of the covariates and the errors. In fact, the $L_2$ loss of the LSE can converge arbitrarily slowly when the independence fails. The key technical ingredient is a new multiplier inequality that gives sharp bounds for the `multiplier empirical process' associated with the LSE. We further give an application to the sparse linear regression model with heavy-tailed covariates and errors to demonstrate the scope of this new inequality.