conquer

Convolution-Type Smoothed Quantile Regression

Estimation and inference for conditional linear quantile regression models using a convolution smoothed approach. In the low-dimensional setting, efficient gradient-based methods are employed for fitting both a single model and a regression process over a quantile range. Normal-based and (multiplier) bootstrap confidence intervals for all slope coefficients are constructed. In high dimensions, the conquer method is complemented with flexible types of penalties (Lasso, elastic-net, group lasso, sparse group lasso, scad and mcp) to deal with complex low-dimensional structures.

Tests Vignettes

Dependencies

Imports/Depends/LinkingTo/Enhances (5)
  • R
  • Rcpp
  • RcppArmadillo >= 0.9.850.1.0
  • Rcpp >= 1.0.3
  • matrixStats
  • Version History