English

Quantile Regression for Large-scale Applications

Data Structures and Algorithms 2014-01-08 v3 Distributed, Parallel, and Cluster Computing Numerical Analysis Machine Learning

Abstract

Quantile regression is a method to estimate the quantiles of the conditional distribution of a response variable, and as such it permits a much more accurate portrayal of the relationship between the response variable and observed covariates than methods such as Least-squares or Least Absolute Deviations regression. It can be expressed as a linear program, and, with appropriate preprocessing, interior-point methods can be used to find a solution for moderately large problems. Dealing with very large problems, \emph(e.g.), involving data up to and beyond the terabyte regime, remains a challenge. Here, we present a randomized algorithm that runs in nearly linear time in the size of the input and that, with constant probability, computes a (1+ϵ)(1+\epsilon) approximate solution to an arbitrary quantile regression problem. As a key step, our algorithm computes a low-distortion subspace-preserving embedding with respect to the loss function of quantile regression. Our empirical evaluation illustrates that our algorithm is competitive with the best previous work on small to medium-sized problems, and that in addition it can be implemented in MapReduce-like environments and applied to terabyte-sized problems.

Keywords

Cite

@article{arxiv.1305.0087,
  title  = {Quantile Regression for Large-scale Applications},
  author = {Jiyan Yang and Xiangrui Meng and Michael W. Mahoney},
  journal= {arXiv preprint arXiv:1305.0087},
  year   = {2014}
}

Comments

35 pages; long version of a paper appearing in the 2013 ICML. Version to appear in the SIAM Journal on Scientific Computing

R2 v1 2026-06-22T00:09:23.551Z