Approximate Bayesian computation with the Wasserstein distance

Abstract

A growing number of generative statistical models do not permit the numerical evaluation of their likelihood functions. Approximate Bayesian computation (ABC) has become a popular approach to overcome this issue, in which one simulates synthetic data sets given parameters and compares summaries of these data sets with the corresponding observed values. We propose to avoid the use of summaries and the ensuing loss of information by instead using the Wasserstein distance between the empirical distributions of the observed and synthetic data. This generalizes the well-known approach of using order statistics within ABC to arbitrary dimensions. We describe how recently developed approximations of the Wasserstein distance allow the method to scale to realistic data sizes, and propose a new distance based on the Hilbert space-filling curve. We provide a theoretical study of the proposed method, describing consistency as the threshold goes to zero while the observations are kept fixed, and concentration properties as the number of observations grows. Various extensions to time series data are discussed. The approach is illustrated on various examples, including univariate and multivariate g-and-k distributions, a toggle switch model from systems biology, a queueing model, and a Lévy-driven stochastic volatility model.

42 pages, 10 figures. Supplementary materials can be found on the first author's webpage. Portions of this work previously appeared as arXiv:1701.05146

References in corpus (8)

• Statistical Aspects of Wasserstein Distances
• Three discussions of the paper "sequential quasi-Monte Carlo sampling", by M. Gerber and N. Chopin
• The Shortlist Method for Fast Computation of the Earth Mover's Distance and Finding Optimal Solutions to Transportation Problems
• Variance bounding and geometric ergodicity of Markov chain Monte Carlo kernels for approximate Bayesian computation
• On Disturbance State-Space Models and the Particle Marginal Metropolis-Hastings Sampler
• On parameter estimation with the Wasserstein distance
• On Bayesian inference for the M/G/1 queue with efficient MCMC sampling
• A rare event approach to high dimensional Approximate Bayesian computation

Cited by in corpus (29)

• Statistical Inference for Generative Models with Maximum Mean Discrepancy
• Black-box Bayesian inference for economic agent-based models
• Minimax Confidence Intervals for the Sliced Wasserstein Distance
• Approximating Bayes in the 21st Century
• ABCpy: A High-Performance Computing Perspective to Approximate Bayesian Computation
• Minimum Stein Discrepancy Estimators
• Batch simulations and uncertainty quantification in Gaussian process surrogate approximate Bayesian computation
• Generalized Posteriors in Approximate Bayesian Computation
• Robust and Efficient Approximate Bayesian Computation: A Minimum Distance Approach
• Generalized Bayesian Likelihood-Free Inference
• On the Wasserstein median of probability measures
• Distance-learning For Approximate Bayesian Computation To Model a Volcanic Eruption
• Projected Statistical Methods for Distributional Data on the Real Line with the Wasserstein Metric
• Implicit Distributional Reinforcement Learning
• On Transportation of Mini-batches: A Hierarchical Approach
• Data-Space Validation of High-Dimensional Models by Comparing Sample Quantiles
• Approximate Bayesian Computation via Classification
• Robust Optimisation Monte Carlo
• Discrepancy-based Inference for Intractable Generative Models using Quasi-Monte Carlo
• Causal Inference on Distribution Functions
• The Fixed Landscape Inference MethOd (flimo): a versatile alternative to Approximate Bayesian Computation, faster by several orders of magnitude
• ABC Learning of Hawkes Processes with Missing or Noisy Event Times
• Posterior asymptotics in Wasserstein metrics on the real line
• Fiducial Matching for the Approximate Posterior: F-ABC
• From Denoising Diffusions to Denoising Markov Models
• Wasserstein Neural Processes
• Dependence properties and Bayesian inference for asymmetric multivariate copulas
• Approximate Bayesian Computation of Bézier Simplices
• Compatibility-based Bayesian inference: asymptotic limits, variational principles, and Wasserstein geometry