Bayesian Computation of Copula-based Multivariate Models with an Application to Doping Detection

The Athlete Biological Passport (ABP) monitors athletes for doping by comparing their biomarker readings against personalized reference ranges. These are estimated via the Bayesian ADAPTIVE model, which analyzes biomarkers separately and ignores their dependence. Building on a recently proposed copula-based multivariate extension of ADAPTIVE, this paper focuses on its computational aspects, introducing a fully joint MCMC algorithm and comparing it with a two-step Bayesian IFM estimator. Simulations show the two procedures perform comparably under a correctly specified copula, but exhibit a genuine trade-off under misspecification: the Bayesian IFM protects marginal inference, while the joint MCMC yields better-calibrated dependence estimates and is markedly faster. The two approaches are tested on a dataset containing ABP profiles of elite athletes, provided by NADO Italia, and collected as part of the project {\it Enhancing the methodological framework for doping detection: from univariate to multivariate testing analysis on Athlete Biological Passport profiles}. Both approaches achieve consistent classification performance, providing anti-doping organizations with two concrete inferential routes for adopting a multivariate ADAPTIVE model, along with guidance on the trade-offs between them.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-18
DOI
https://doi.org/10.48550/arxiv.2611.24801
Primary Topic
Doping in Sports
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Bayesian Computation of Copula-based Multivariate Models with an Application to Doping Detection

Brunero Liseo, Nina Deliu, Emanuele Masillo, Eva Pigi Viltanioti
arXiv (Cornell University)
Doping in Sports
preprint

Bayesian Computation of Copula-based Multivariate Models with an Application to Doping Detection

Brunero Liseo, Nina Deliu, Emanuele Masillo, Eva Pigi Viltanioti
preprint en

Abstract

The Athlete Biological Passport (ABP) monitors athletes for doping by comparing their biomarker readings against personalized reference ranges. These are estimated via the Bayesian ADAPTIVE model, which analyzes biomarkers separately and ignores their dependence. Building on a recently proposed copula-based multivariate extension of ADAPTIVE, this paper focuses on its computational aspects, introducing a fully joint MCMC algorithm and comparing it with a two-step Bayesian IFM estimator. Simulations show the two procedures perform comparably under a correctly specified copula, but exhibit a genuine trade-off under misspecification: the Bayesian IFM protects marginal inference, while the joint MCMC yields better-calibrated dependence estimates and is markedly faster. The two approaches are tested on a dataset containing ABP profiles of elite athletes, provided by NADO Italia, and collected as part of the project {\it Enhancing the methodological framework for doping detection: from univariate to multivariate testing analysis on Athlete Biological Passport profiles}. Both approaches achieve consistent classification performance, providing anti-doping organizations with two concrete inferential routes for adopting a multivariate ADAPTIVE model, along with guidance on the trade-offs between them.

arXiv (Cornell University)
Doping in Sports
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon