B1325
Title: Quantile martingale difference divergence for dimension reduction
Authors: Chung Eun Lee - Baruch College (United States) [presenting]
Haileab Hilafu - University of Tennessee (United States)
Abstract: The aim is to reduce the dimension of predictors by considering the central quantile subspace. To do so, we use a metric, the quantile martingale difference divergence which measures the quantile dependence of a scalar response variable and a vector of predictors. The proposed dimension-reduction method does not involve user-chosen parameters and does not assume a parametric model, making them simple to implement. Extensive simulations and a real-data illustration are provided to demonstrate the usefulness of the proposed method, which are shown to yield competitive finite-sample performance.