Coverage Error Optimal Confidence Intervals for Local Polynomial Regression

DOI10.3150/21-BEJ1445MaRDI QIDQ6305013arXivFDO

Authors Sebastian Calonico, Matias D. Cattaneo, Max H. Farrell

Publication date 3 August 2018

Published in arXiv

Full work available at URL https://doi.org/10.3150/21-BEJ1445

Mathematics Subject Classification ID

Statistical distribution theory (62Exx) Nonparametric inference (62Gxx) Probabilistic methods, stochastic differential equations (65Cxx)

arXiv classification

econ.EMmath.STstat.TH

Abstract: This paper studies higher-order inference properties of nonparametric local polynomial regression methods under random sampling. We prove Edgeworth expansions for $t$ statistics and coverage error expansions for interval estimators that (i) hold uniformly in the data generating process, (ii) allow for the uniform kernel, and (iii) cover estimation of derivatives of the regression function. The terms of the higher-order expansions, and their associated rates as a function of the sample size and bandwidth sequence, depend on the smoothness of the population regression function, the smoothness exploited by the inference procedure, and on whether the evaluation point is in the interior or on the boundary of the support. We prove that robust bias corrected confidence intervals have the fastest coverage error decay rates in all cases, and we use our results to deliver novel, inference-optimal bandwidth selectors. The main methodological results are implemented in companion extsf{R} and extsf{Stata} software packages.

This page was built for publication: Coverage Error Optimal Confidence Intervals for Local Polynomial Regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6305013)