Inference for concave distribution functions under measurement error

TL;DR

This paper introduces a new estimator for concave distribution functions under measurement error, proves its consistency and convergence, and develops a bootstrap-based test for the concavity assumption, with theoretical justification and simulation validation.

Contribution

It presents a novel estimator based on the least concave majorant of a deconvolution estimator and a bootstrap test for concavity, with rigorous theoretical backing.

Findings

Estimator is uniformly consistent and converges at a √n rate.

Bootstrap test correctly calibrates the null hypothesis and has high power.

Simulation studies demonstrate good finite-sample performance.

Abstract

We propose an estimator of a concave cumulative distribution function under the measurement error model, where the non-negative variables of interest are perturbed by additive independent random noise. The estimator is defined as the least concave majorant on the positive half-line of the deconvolution estimator of the distribution function. We show its uniform consistency and its square root convergence in law in $\ell_\infty(\mathbb R)$. To assess the validity of the concavity assumption, we construct a test for the nonparametric null hypothesis that the distribution function is concave on the positive half-line, against the alternative that it is not. We calibrate the test using bootstrap methods. The theoretical justification for calibration led us to establish a bootstrap version of Theorem 1 in S\"ohl and Trabs (2012), a Donsker-type result from which we obtain, as a special case, the limiting behavior of the deconvolution estimator of the distribution function in a bootstrap setting with measurement error. Combining this Donsker-type theorem with the functional delta method, we show that the test statistic and its bootstrap version have the same limiting distribution under the null hypothesis, whereas under the alternative, the bootstrap statistic is stochastically smaller. Consequently, the power of the test tends to one, for any fixed alternative, as the sample size tends to infinity. In addition to the theoretical results for the estimator and the test, we investigate their finite-sample performance in simulation studies.