Stratified regression-based variance reduction approach for weak approximation schemes

TL;DR

This paper introduces a stratified regression-based variance reduction method to improve the efficiency of weak approximation schemes, demonstrating significant variance reduction through numerical examples.

Contribution

It proposes a novel modification combining stratification with regression-based variance reduction, enhancing performance over existing methods.

Findings

Significant variance reduction achieved in numerical tests

Improved efficiency of weak approximation schemes

Demonstrated effectiveness across multiple examples

Abstract

In this paper we suggest a modification of the regression-based variance reduction approach recently proposed in Belomestny et al. This modification is based on the stratification technique and allows for a further significant variance reduction. The performance of the proposed approach is illustrated by several numerical examples.