Defining replicability of prediction rules

DOI10.1214/23-STS891arXiv2305.01518OpenAlexW4388460646MaRDI QIDQ6145147FDOQ6145147

Authors: Giovanni Parmigiani 

Publication date: 30 January 2024

Published in: Statistical Science (Search for Journal in Brave)

Abstract: In this article I propose an approach for defining replicability for prediction rules. Motivated by a recent NAS report, I start from the perspective that replicability is obtaining consistent results across studies suitable to address the same prediction question, each of which has obtained its own data. I then discuss concept and issues in defining key elements of this statement. I focus specifically on the meaning of "consistent results" in typical utilization contexts, and propose a multi-agent framework for defining replicability, in which agents are neither partners nor adversaries. I recover some of the prevalent practical approaches as special cases. I hope to provide guidance for a more systematic assessment of replicability in machine learning.

Full work available at URL: https://arxiv.org/abs/2305.01518

zbMATH Keywords

predictionreplicabilitydecision theory

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• Title not available (Why is that?)
• Statistical modeling: The two cultures. (With comments and a rejoinder).
• Fairness through awareness
• Veridical data science
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Stacked regressions
• Thomas Bayes's Bayesian Inference
• Hierarchical resampling for bagging in multistudy prediction with applications to human neurochemical sensing
• Training replicable predictors in multiple studies
• Modeling between-study heterogeneity for improved replicability in gene signature selection and clinical prediction
• Bayesian nonparametric cross-study validation of prediction methods
• Tracking cross-validated estimates of prediction error as studies accumulate
• Integration of survival data from multiple studies

This page was built for publication: Defining replicability of prediction rules

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6145147)