Conditional inferential models: combining information for prior-free probabilistic inference

Abstract

The inferential model (IM) framework provides valid prior-free probabilistic inference by focusing on predicting unobserved auxiliary variables. But, efficient IM-based inference can be challenging when the auxiliary variable is of higher dimension than the parameter. Here we show that features of the auxiliary variable are often fully observed and, in such cases, a simultaneous dimension reduction and information aggregation can be achieved by conditioning. This proposed conditioning strategy leads to efficient IM inference, and casts new light on Fisher's notions of sufficiency, conditioning, and also Bayesian inference. A differential equation-driven selection of a conditional association is developed, and validity of the conditional IM is proved under some conditions. For problems that do not admit a valid conditional IM of the standard form, we propose a more flexible class of conditional IMs based on localization. Examples of local conditional IMs in a bivariate normal model and a normal variance components model are also given.

27 pages, 2 tables, 4 figures. Proposition 1 and its corollary are not included in the published version

References in corpus (6)

• The formal definition of reference priors
• Inferential models: A framework for prior-free posterior probabilistic inference
• Is Bayes Posterior just Quick and Dirty Confidence?
• Marginal inferential models: prior-free probabilistic inference on interest parameters
• Generalized fiducial inference for normal linear mixed models
• Second order ancillary: A differential view from continuity

Cited by in corpus (25)

• Inferential models: A framework for prior-free posterior probabilistic inference
• Marginal inferential models: prior-free probabilistic inference on interest parameters
• False confidence, non-additive beliefs, and valid statistical inference
• Plausibility functions and exact frequentist inference
• On an inferential model construction using generalized associations
• A note on p-values interpreted as plausibilities
• Frameworks for prior-free posterior probabilistic inference
• Validity, consonant plausibility measures, and conformal prediction
• Valid inferential models for prediction in supervised learning problems
• Exact prior-free probabilistic inference on the heritability coefficient in a linear mixed model
• Discussion: Foundations of Statistical Inference, Revisited
• Higher order asymptotics of Generalized Fiducial Distribution
• A mathematical characterization of confidence as valid belief
• Inferential models and possibility measures
• Possibilistic inferential models: a review
• Discussion of "On the Birnbaum Argument for the Strong Likelihood Principle"
• Generalized inferential models for meta-analyses based on few studies
• Computationally efficient variational-like approximations of possibilistic inferential models
• Valid uncertainty quantification about the model in a linear regression setting
• Confidence Distribution and Distribution Estimation for Modern Statistical Inference
• Partial Conditioning for Inference of Many-Normal-Means with Hölder Constraints
• A fiducial approach to nonparametric deconvolution problem: discrete case
• Exact prior-free probabilistic inference in a class of non-regular models
• A Geometric Perspective on Bayesian and Generalized Fiducial Inference
• A simple recipe for making accurate parametric inference in finite sample