Quantile regression under memory constraint (Q2284373)

The authors study the inference problem in quantile regression (QR) for a large sample size \(n\), and describe a general divide-and-conquer algorithm for statistical estimation. The entire \(n\) data set is divided into \(N\) subsets with equal size \(m\) \((N = n/m)\). Based on the linear estimator for quantile regression, a divide-and-conquer linear estimator for quantile regression is introduced by the authors under memory constraints.