Rank-normalization, folding, and localization: An improved $\hat{R}$ for assessing convergence of MCMC

Abstract

Markov chain Monte Carlo is a key computational tool in Bayesian statistics, but it can be challenging to monitor the convergence of an iterative stochastic algorithm. In this paper we show that the convergence diagnostic $\hat{R}$ of Gelman and Rubin (1992) has serious flaws. Traditional $\hat{R}$ will fail to correctly diagnose convergence failures when the chain has a heavy tail or when the variance varies across the chains. In this paper we propose an alternative rank-based diagnostic that fixes these problems. We also introduce a collection of quantile-based local efficiency measures, along with a practical approach for computing Monte Carlo error estimates for quantiles. We suggest that common trace plots should be replaced with rank plots from multiple chains. Finally, we give recommendations for how these methods should be used in practice.

Two small fixes. Published in Bayesian analysis https://doi.org/10.1214/20-BA1221

Cited by in corpus (116)

• The NANOGrav 15-year Data Set: Evidence for a Gravitational-Wave Background
• First Sagittarius A* Event Horizon Telescope Results. IV. Variability, Morphology, and Black Hole Mass
• Early Release Science of the exoplanet WASP-39b with JWST NIRISS
• The Effects of Generative AI on Design Fixation and Divergent Thinking
• Full-shape cosmology analysis of SDSS-III BOSS galaxy power spectrum using emulator-based halo model: a $5\%$ determination of ${\sigma }_8$
• Detecting and diagnosing prior and likelihood sensitivity with power-scaling
• Graphical Test for Discrete Uniformity and its Applications in Goodness of Fit Evaluation and Multiple Sample Comparison
• The NANOGrav 15-year Gravitational-Wave Background Methods
• Simulation-Based Calibration Checking for Bayesian Computation: The Choice of Test Quantities Shapes Sensitivity
• Sequential Monte Carlo for Sampling Balanced and Compact Redistricting Plans
• Bayesian neural networks via MCMC: a Python-based tutorial
• First semi-empirical test of the white dwarf mass-radius relationship using a single white dwarf via astrometric microlensing
• Quantum Magic and Multi-Partite Entanglement in the Structure of Nuclei
• Testing $\Lambda$-Free f(Q) Cosmology
• Constraining the SN Ia Host Galaxy Dust Law Distribution and Mass Step: Hierarchical BayeSN Analysis of Optical and Near-Infrared Light Curves
• Bayesian parameter estimation in $\chi$EFT using Hamiltonian Monte Carlo
• Uncertainty Quantification and Propagation in Surrogate-based Bayesian Inference
• Post-Processing of MCMC
• Disentangling Multiple Stochastic Gravitational Wave Background Sources in PTA Datasets
• Impact of the Euro 2020 championship on the spread of COVID-19
• Investigating the turbulent hot gas in X-COP galaxy clusters
• Some models are useful, but how do we know which ones? Towards a unified Bayesian model taxonomy
• MAXI J1848-015: The First Detection of Relativistically Moving Outflows from a Globular Cluster X-ray Binary
• Horseshoe Prior Bayesian Quantile Regression
• Probing the Mechanism of Neutrinoless Double-Beta Decay in Multiple Isotopes
• A fully Bayesian sparse polynomial chaos expansion approach with joint priors on the coefficients and global selection of terms
• On the hierarchical Bayesian modelling of frequency response functions
• Using Rest-Frame Optical and NIR Data from the RAISIN Survey to Explore the Redshift Evolution of Dust Laws in SN Ia Host Galaxies
• Exoplanet atmosphere evolution: emulation with neural networks
• CUQIpy: I. Computational uncertainty quantification for inverse problems in Python
• The G-Wishart Weighted Proposal Algorithm: Efficient Posterior Computation for Gaussian Graphical Models
• Advice on describing Bayesian analysis of neutron and X-ray reflectometry
• Enhanced monitoring of atmospheric methane from space over the Permian basin with hierarchical Bayesian inference
• A comparison of Bayesian sampling algorithms for high-dimensional particle physics and cosmology applications
• Grid-Based Atmospheric Retrievals for Reflected-Light Spectra of Exoplanets using PSGnest
• Uncertainty in Bayesian Leave-One-Out Cross-Validation Based Model Comparison
• GOES GLM, Biased Bolides, and Debiased Distributions
• Network reconstruction via the minimum description length principle
• Discriminating among cosmological models by data-driven methods
• Tracking the Transmission Dynamics of COVID-19 with a Time-Varying Coefficient State-Space Model
• Shock-cooling Constraints via Early-time Observations of the Type IIb SN 2022hnt
• Identifying supermassive black hole recoil in elliptical galaxies
• Mapping the prevalence of cancer risk factors at the small area level in Australia
• Modelling variability power spectra of active galaxies from irregular time series
• Almanac: MCMC-based signal extraction of power spectra and maps on the sphere
• Estimating fine age structure and time trends in human contact patterns from coarse contact data: the Bayesian rate consistency model
• Observations of scattered light from exoplanet atmospheres
• A Large Double-ring Disk around the Taurus M Dwarf J04124068+2438157
• Using a Bayesian-Inference Approach to Calibrating Models for Simulation in Robotics
• Approximately-symmetric neural networks for quantum spin liquids
• DAmodel: Hierarchical Bayesian Modelling of DA White Dwarfs for Spectrophotometric Calibration
• Multilevel network meta-regression for general likelihoods: synthesis of individual and aggregate data with applications to survival analysis
• A Differentiable, End-to-End Forward Model for 21 cm Cosmology: Estimating the Foreground, Instrument, and Signal Joint Posterior
• Towards quantum gravity with neural networks: Solving the quantum Hamilton constraint of U(1) BF theory
• Modeling Magnetic Flux Emergence in Bipolar Active Regions
• Bayesian approach for modeling solar active region global magnetic parameters
• Toward Filling a Critical Knowledge Gap: Charting the Interactions of Age with Task and Visualization
• Requirements Quality Research Artifacts: Recovery, Analysis, and Management Guideline
• Solar-like oscillations in $\gamma$ Cephei A as seen through SONG and TESS
• Makeham Mortality Models as Mixtures
• A rapidly mixing Markov chain from any gapped quantum many-body system
• Quantum-enhanced Markov Chain Monte Carlo for systems larger than your Quantum Computer
• jaxspec : a fast and robust Python library for X-ray spectral fitting
• Truncated Partial-Wave Analysis for $\eta$-photoproduction observables via Bayesian Statistics
• Adaptive multi-stage integration schemes for Hamiltonian Monte Carlo
• Combining chains of Bayesian models with Markov melding
• Projection predictive variable selection for discrete response families with finite support
• Understanding and predicting cadence effects in the characterization of exoplanet transits
• An importance sampling approach for reliable and efficient inference in Bayesian ordinary differential equation models
• Precise determination of circumstellar disk lifetimes: Disk evolution in a single star-forming region
• Scalable and Robust Multiband Modeling of AGN Light Curves in Rubin-LSST
• Neural Quantum State Study of Fracton Models
• DGP-LVM: Derivative Gaussian process latent variable models
• Big Bang Nucleosynthesis constraints on space-time noncommutativity
• Statistical Models for Repeated Categorical Ratings: The R Package rater
• Generalized Decomposition Priors on R2
• stanhf: HistFactory models in the probabilistic programming language Stan
• Inferring flavor mixtures in multijet events
• Toward mapping turbulence in the intracluster medium III. Constraints on the turbulent power spectrum with Athena/X-IFU
• Bayesian approach to equipartition estimation of magnetic field strength
• Modeling the $R$-ratio and hadronic contributions to $g-2$ with a Treed Gaussian Process
• On decision-theoretic model assessment for structural deterioration monitoring
• True Masses using RV data with Hipparcos and Gaia Astrometry
• A differentiable N-body code for transit timing and dynamical modelling -- II. Photodynamics
• Modeling racial/ethnic differences in COVID-19 incidence with covariates subject to non-random missingness
• A Matched Survey for the Enigmatic Low Radio Frequency Transient ILT J225347+862146
• Symmetry-driven embedding of networks in hyperbolic space
• Gaussian distributional structural equation models: A framework for modeling latent heteroscedasticity
• Mathematical modelling and uncertainty quantification for analysis of biphasic coral reef recovery patterns
• Bayesian inference methodology to characterize the dust emissivity at far-infrared and submillimeter frequencies
• Benchmarking Diffusion Annealing-Based Bayesian Inverse Problem Solvers
• Bayesian random-effects meta-analysis of aggregate data on clinical events
• Prediction can be safely used as a proxy for explanation in causally consistent Bayesian generalized linear models
• Confirmation of the hot super-Neptune TOI-672 b with NIRPS and HARPS and Insights into the Neptunian desert around M dwarfs
• New Metrics for Identifying Variables and Transients in Large Astronomical Surveys
• Statistical Inferences and Predictions for Areal Data and Spatial Data Fusion with Hausdorff--Gaussian Processes
• Topological models to infer multiphase interstellar medium properties
• In situ quantum verification of polarization-stabilized optical channels
• Bayesian analysis of diffusion-driven multi-type epidemic models with application to COVID-19
• Price Optimization Combining Conjoint Data and Purchase History: A Causal Modeling Approach
• Modeling of silver transport in cubic SiC: Integrating molecular dynamics, bounds averaging, and uncertainty quantification
• Knowledge Distillation of a Protein Language Model Yields a Foundational Implicit Solvent Model
• Random time-shift approximation enables hierarchical Bayesian inference of mechanistic within-host viral dynamics models on large datasets
• Hilbert space methods for approximating multi-output latent variable Gaussian processes
• A Framework for the Bayesian Calibration of Complex and Data-Scarce Models in Applied Sciences
• FlowSN: Neural Simulation-Based Inference under Realistic Selection Effects applied to Supernova Cosmology
• KMT-2025-BLG-1314 and KMT-2025-BLG-1392: two microlensing planetary/brown-dwarf candidates analyzed with differentiable code
• Bayesian Inference of Nonlinear Malaria Dynamics in Ghana via an Ensemble Markov Chain Monte Carlo Sampler
• Hidden Markov modelling of spatio-temporal dynamics of measles in 1750-1850 Finland
• Bayesian generalized method of moments applied to pseudo-observations in survival analysis
• An Agnostic Approach to Building Empirical Type Ia Supernova Light Curves: Evidence for Intrinsic Chromatic Flux Variation Using Nearby Supernova Factory Data
• No Silver Bullets: Why Understanding Software Cycle Time is Messy, Not Magic
• Bring the noise: exact inference from noisy simulations in collider physics
• Stellar Cores Live Long and Prosper in Cuspy Dark Matter Halos
• Inter-turbine Modelling of Wind-Farm Power using Multi-task Learning
• Optimizing the dynamical preparation of quantum spin lakes on the ruby lattice