Model criticism in latent space

DOI10.1214/18-BA1124MaRDI QIDQ2316980zbMATH OpenOpenAlexFDO

Authors Sohan Seth, Iain Murray, Christopher K. I. Williams

Publication date 7 August 2019

Published in Bayesian Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1711.04674, https://projecteuclid.org/euclid.ba/1560240024

zbMATH Keywords

factor analysisGaussian processeslatent variable modelslinear dynamical systemsmodel criticism

Mathematics Subject Classification ID

Dynamical aspects of statistical mechanics (37A60) Gaussian processes (60G15) Foundations and philosophical topics in statistics (62A01) Theory of statistical experiments (62B15) Factor analysis and principal components; correspondence analysis (62H25)

Abstract: Model criticism is usually carried out by assessing if replicated data generated under the fitted model looks similar to the observed data, see e.g. Gelman, Carlin, Stern, and Rubin [2004, p. 165]. This paper presents a method for latent variable models by pulling back the data into the space of latent variables, and carrying out model criticism in that space. Making use of a model's structure enables a more direct assessment of the assumptions made in the prior and likelihood. We demonstrate the method with examples of model criticism in latent space applied to factor analysis, linear dynamical systems and Gaussian processes.

Recommendations

• Diagnostic Measures for Model Criticism
• Bayesian comparison of latent variable models: conditional versus marginal likelihoods
• SOME REMARKS ON MODEL CRITICISM
• scientific article; zbMATH DE number 1522705
• scientific article; zbMATH DE number 1489818

Cites work

• A Fast Learning Algorithm for Deep Belief Nets
• A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity
• A Simple Test for Heteroscedasticity and Random Coefficient Variation
• Bayesian model assessment using pivotal quantities
• Bayesianly justifiable and relevant frequency calculations for the applied statistician
• Gaussian processes for machine learning.
• Goodness-of-fit diagnostics for Bayesian hierarchical models
• GPstuff: Bayesian modeling with Gaussian processes
• scientific article; zbMATH DE number 947414 (Why is no real title available?)
• scientific article; zbMATH DE number 3994822 (Why is no real title available?)
• scientific article; zbMATH DE number 1911984 (Why is no real title available?)
• scientific article; zbMATH DE number 3059918 (Why is no real title available?)
• P Values for Composite Null Models
• Probabilistic Feature Analysis of Facial Perception of Emotions
• Sampling and Bayes' Inference in Scientific Modelling and Robustness

Cited in

(1)

• Using prior expansions for prior-data conflict checking

Describes a project that uses

Uses Software

• BSDS
• JAGS
• darch
• BayesDA
• GPstuff

This page was built for publication: Model criticism in latent space

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2316980)