Optimal Nonparametric Inference with Two-Scale Distributional Nearest Neighbors

This paper provide an in-depth technical analysis of the distribution nearest neighbors (DNN) method, based on which we suggest a bias reduction approach for the DNN estimator by linearly combining two DNN estimators with different subsampling scales, resulting in the novel two-scale DNN (TDNN) estimator.

July 2022 · Emre Demirkayaa, Yingying Fan, Lan Gao, Jinchi Lv, Patrick Vossler, Jingbo Wang