High-dimensional varying index coefficient quantile regression model
share
Statistical learning evolves quickly with more and more sophisticated models proposed to incorporate the complicated data structure from modern scientific and business problems. Varying index coefficient models extend varying coefficient models and single index models, becoming the latest state-of-the-art for semiparametric regression. This new class of models offers greater flexibility to characterize complicated nonlinear interaction effects in regression analysis. To safeguard against outliers and extreme observations, we consider a robust quantile regression approach to estimate the model parameters in this paper. High-dimensional loading parameters are allowed in our development under reasonable theoretical conditions. In addition, we propose a regularized estimation procedure to choose between linear and non-linear forms for interaction terms. We can simultaneously select significant non-zero loading parameters and identify linear functions in varying index coefficient models, in addition to estimate all the parametric and nonparametric components consistently. Under technical assumptions, we show that the proposed procedure is consistent in variable selection as well as in linear function identification, and the proposed parameter estimation enjoys the oracle property. Extensive simulation studies are carried out to assess the finite sample performance of the proposed method. We illustrate our methods with an environmental health data example.
READ FULL TEXT