Search:
Match:
2 results
Research Paper#Bayesian Statistics, Survival Analysis, MCMC, Mixture Models🔬 ResearchAnalyzed: Jan 3, 2026 18:39

Improving Bayesian Profile Regression for Survival Analysis

Published:Dec 29, 2025 16:11
1 min read
ArXiv

Analysis

This paper addresses the instability issues in Bayesian profile regression mixture models (BPRM) used for assessing health risks in multi-exposed populations. It focuses on improving the MCMC algorithm to avoid local modes and comparing post-treatment procedures to stabilize clustering results. The research is relevant to fields like radiation epidemiology and offers practical guidelines for using these models.
Reference

“The paper proposes improvements to MCMC algorithms and compares post-processing methods to stabilize the results of Bayesian profile regression mixture models.”

PermalinkArXiv
Research#Causal Inference🔬 ResearchAnalyzed: Jan 10, 2026 08:58

PIPCFR: Estimating Treatment Effects with Post-Treatment Variables

Published:Dec 21, 2025 13:57
1 min read
ArXiv

Analysis

This ArXiv paper introduces a novel method (PIPCFR) for estimating individual treatment effects. The focus on handling post-treatment variables is particularly relevant in causal inference, where traditional methods can be biased.
Reference

“PIPCFR: Pseudo-outcome Imputation with Post-treatment Variables for Individual Treatment Effect Estimation”

PermalinkArXiv