Minimally Biased Nonparametric Regression and Autoregression

Authors

  • Timothy L. McMurry DePaul University
  • Dimitris N. Politis University of California

DOI:

https://doi.org/10.57805/revstat.v6i2.61

Keywords:

nonparametric regression, autoregression, Fourier transform

Abstract

A nonparametric regression estimator is introduced which adapts to the smoothness of the unknown function being estimated. This property allows the new estimator to automatically achieve minimal bias over a large class of locally smooth functions without changing the rate at which the variance converges. Optimal convergence rates are shown to hold for both i.i.d. data and autoregressive processes satisfying strong mixing conditions.

Published

2008-06-24

How to Cite

L. McMurry , T., & N. Politis , D. (2008). Minimally Biased Nonparametric Regression and Autoregression. REVSTAT-Statistical Journal, 6(2), 123–150. https://doi.org/10.57805/revstat.v6i2.61