Inference for local parameters in convexity constrained models

TL;DR

This paper develops a new pivotal inference method for local parameters in convex regression models, enabling accurate, tuning-free confidence intervals without needing to estimate second derivatives or error variance.

Contribution

It introduces locally normalized errors (LNEs) with universal limiting distributions for local functionals in convex regression, facilitating practical inference.

Findings

LNEs have asymptotically pivotal limiting distributions.

Constructs simple, tuning-free confidence intervals for function values and derivatives.

Extends the theory to other convexity-constrained models like log-concave density estimation.

Abstract

We consider the problem of inference for local parameters of a convex regression function $f_0: [0,1] \to \mathbb{R}$ based on observations from a standard nonparametric regression model, using the convex least squares estimator (LSE) $\widehat{f}_n$. For $x_0 \in (0,1)$, the local parameters include the pointwise function value $f_0(x_0)$, the pointwise derivative $f_0'(x_0)$, and the anti-mode (i.e., the smallest minimizer) of $f_0$. The existing limiting distribution of the estimation error $(\widehat{f}_n(x_0) - f_0(x_0), \widehat{f}_n'(x_0) - f_0'(x_0) )$ depends on the unknown second derivative $f_0''(x_0)$, and is therefore not directly applicable for inference. To circumvent this impasse, we show that the following locally normalized errors (LNEs) enjoy pivotal limiting behavior: Let $[\widehat{u}(x_0), \widehat{v}(x_0)]$ be the maximal interval containing $x_0$ where $\widehat{f}_n$ is linear. Then, under standard conditions, $$\binom{ \sqrt{n(\widehat{v}(x_0)-\widehat{u}(x_0))}(\widehat{f}_n(x_0)-f_0(x_0)) }{ \sqrt{n(\widehat{v}(x_0)-\widehat{u}(x_0))^3}(\widehat{f}_n'(x_0)-f_0'(x_0))} \rightsquigarrow \sigma \cdot \binom{\mathbb{L}^{(0)}_2}{\mathbb{L}^{(1)}_2},$$ where $n$ is the sample size, $\sigma$ is the standard deviation of the errors, and $\mathbb{L}^{(0)}_2, \mathbb{L}^{(1)}_2$ are universal random variables. This asymptotically pivotal LNE theory instantly yields a simple tuning-free procedure for constructing CIs with asymptotically exact coverage and optimal length for $f_0(x_0)$ and $f_0'(x_0)$. We also construct an asymptotically pivotal LNE for the anti-mode of $f_0$, and its limiting distribution does not even depend on $\sigma$. These asymptotically pivotal LNE theories are further extended to other convexity/concavity constrained models (e.g., log-concave density estimation) for which a limit distribution theory is available for problem-specific estimators.