This paper establishes the asymptotic validity of process-level bootstrap combinations for stationary long-memory (ARFIMA) processes. We analyze simple averages, robust operators, and adaptive weighting schemes, proving their consistency in the Mallows distance of order two. Crucially, we derive a convergence rate bound determined by the slowest component method. This result provides a theoretical safeguard, guaranteeing that aggregation preserves asymptotic validity while mitigating the sensitivity to tuning parameters typical of individual resampling methods.

Asymptotic Validity of Bootstrap Combinations for Long-Memory Time Series

Margherita Palomba;Luisa Bisaglia
2026

Abstract

This paper establishes the asymptotic validity of process-level bootstrap combinations for stationary long-memory (ARFIMA) processes. We analyze simple averages, robust operators, and adaptive weighting schemes, proving their consistency in the Mallows distance of order two. Crucially, we derive a convergence rate bound determined by the slowest component method. This result provides a theoretical safeguard, guaranteeing that aggregation preserves asymptotic validity while mitigating the sensitivity to tuning parameters typical of individual resampling methods.
2026
Statistical Science: From Theory to Applied Research IV
SIS-FENStatS 2026 2026
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11577/3623610
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
  • OpenAlex ND
social impact