Interval estimation for messy observational data

DOI10.1214/09-STS305zbMATH Open1329.62133arXiv1010.0306OpenAlexW2090591864MaRDI QIDQ907959FDOQ907959

Authors: Paul Gustafson, Sander Greenland 

Publication date: 2 February 2016

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

Abstract: We review some aspects of Bayesian and frequentist interval estimation, focusing first on their relative strengths and weaknesses when used in "clean" or "textbook" contexts. We then turn attention to observational-data situations which are "messy," where modeling that acknowledges the limitations of study design and data collection leads to nonidentifiability. We argue, via a series of examples, that Bayesian interval estimation is an attractive way to proceed in this context even for frequentists, because it can be supplied with a diagnostic in the form of a calibration-sensitivity simulation analysis. We illustrate the basis for this approach in a series of theoretical considerations, simulations and an application to a study of silica exposure and lung cancer.

Full work available at URL: https://arxiv.org/abs/1010.0306

Recommendations

• Frequentist and Bayesian approaches for interval-censored data
• Nonparametric Bayesian estimation based on interval censored data
• Bayesian confidence intervals of proportion with misclassified binary data
• Frequentist accuracy of Bayesian estimates
• scientific article; zbMATH DE number 1067781

zbMATH Keywords

Bayesian analysisbiasidentifiabilityconfoundingobservational studiesepidemiologyinterval coveragehierarchical prior

Mathematics Subject Classification ID

Bayesian inference (62F15) Parametric tolerance and confidence regions (62F25)

Cites Work

• Sampling and Bayes' Inference in Scientific Modelling and Robustness
• Title not available (Why is that?)
• Bayesianly justifiable and relevant frequency calculations for the applied statistician
• Multiple-Bias Modelling for Analysis of Observational Data
• Empirical Bayes Analysis of a Microarray Experiment
• Title not available (Why is that?)
• Ignorance and uncertainty regions as inferential tools in a sensitivity analysis
• Parametric Empirical Bayes Inference: Theory and Applications
• Confidence Intervals for Partially Identified Parameters
• The interplay of Bayesian and frequentist analysis
• Title not available (Why is that?)
• In-season prediction of batting averages: a field test of empirical Bayes and Bayes methodol\-ogies
• Frequentist probability and frequentist statistics
• Bayesian approaches to modeling the conditional dependence between multiple diagnostic tests
• On the informativeness and accuracy of pooled testing in estimating prevalence of a rare disease: Application to HIV screening
• On model expansion, model contraction, identifiability and prior information: two illustrative scenarios involving mismeasured variables
• Case-control analysis with partial knowledge of exposure misclassification probabilities
• The Impact of Prior Distributions for Uncontrolled Confounding and Response Bias
• Incorporating prior beliefs about selection bias into the analysis of randomized trials with missing outcomes
• The Performance of Random Coefficient Regression in Accounting for Residual Confounding
• False Models and Post-Data Model Construction
• Relaxation penalties and priors for plausible modeling of nonidentified bias sources
• Likelihood-based confidence sets for partially identified parameters
• The optimal confidence region for a random parameter
• Bayes and Admissible Set Estimation
• Efficiently Simulating the Coverage Properties of Interval Estimates

Cited In (11)

• Partially identified prevalence estimation under misclassification using the kappa coefficient
• Bayes Factors and Posterior Estimation: Two Sides of the Very Same Coin
• Bayesian adjustment for preferential testing in estimating infection fatality rates, as motivated by the COVID-19 pandemic
• Parameter restrictions for the sake of identification: is there utility in asserting that perhaps a restriction holds?
• Using Secondary Outcomes to Sharpen Inference in Randomized Experiments With Noncompliance
• A note on variability of interval data
• Comment: The need for syncretism in applied statistics
• Relaxation penalties and priors for plausible modeling of nonidentified bias sources
• A Bayesian approach to improved estimation of causal effect predictiveness for a principal surrogate endpoint
• On the behaviour of Bayesian credible intervals in partially identified models
• A Bayesian approach for interval data analysis

Uses Software

• EBarrays

This page was built for publication: Interval estimation for messy observational data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q907959)