Conditional quantile sequential estimation for stochastic codes
arXiv:1508.06505
The paper introduces a sequential algorithm that estimates conditional quantiles for stochastic computer codes with vector inputs, using k‑nearest neighbor smoothing inside a Robbins‑Monro procedure, and provides convergence guarantees and non‑asymptotic error rates.
Abstract
We propose and analyze an algorithm for the sequential estimation of a conditional quantile in the context of real stochastic codes with vectorvalued inputs. Our algorithm is based on k-nearest neighbors smoothing within a Robbins-Monro estimator. We discuss the convergence of the algorithm under some conditions on the stochastic code. We provide non-asymptotic rates of convergence of the mean squared error and we discuss the tuning of the algorithm's parameters.