Skip to main navigation Skip to search Skip to main content

Adaptive stratified Monte Carlo using decision trees

  • Tsinghua University
  • University of Bristol

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

It has been known for a long time that stratification is one possible strategy to obtain higher convergence rates for the Monte Carlo estimation of integrals over the hypercube [0,1]s of dimension s. However, stratified estimators such as Haber’s are not practical as s grows, as they require O(ks) evaluations for some k≥2. We propose an adaptive stratification strategy, where the strata are derived from a decision tree applied to a preliminary sample. We show that this strategy leads to higher convergence rates, that is the corresponding estimators converge at rate O(N-1/2-r) for some r>0 for certain classes of functions. Empirically, we show through numerical experiments that the method may improve on standard Monte Carlo even when s is large.

Original languageEnglish
Article number193
JournalStatistics and Computing
Volume35
Issue number6
DOIs
Publication statusPublished - 1 Dec 2025

Keywords

  • Monte Carlo
  • Stratification
  • control variates
  • decision trees

Fingerprint

Dive into the research topics of 'Adaptive stratified Monte Carlo using decision trees'. Together they form a unique fingerprint.

Cite this