# Some cautionary tales about Bayesian predictive inference

arXiv: https://arxiv.org/abs/2607.19206 · HTML: https://mlsanity.com/paper/2607.19206/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 19 references that are in the map.
- First posted on arXiv 21 Jul 2026.

## Builds on (most cited first, 6 of 6)

- [Bayesian Causal Inference: A Critical Review](https://mlsanity.com/paper/2206.15460/) (2022)
- [On recursive Bayesian predictive distributions](https://mlsanity.com/paper/1508.07448/) (2015)
- [Quasi-Bayes properties of a recursive procedure for mixtures](https://mlsanity.com/paper/1902.10708/) (2019)
- [A probabilistic view on predictive constructions for Bayesian learning](https://mlsanity.com/paper/2208.06785/) (2022)
- [Posterior inference via Hill's prediction model](https://mlsanity.com/paper/2603.20071/) (2026)
- [Posterior uncertainty for kernel density estimates](https://mlsanity.com/paper/2607.03927/) (2026)

## Abstract (arXiv)

Two misunderstandings, frequently arising in Bayesian predictive inference, are discussed. The first deals with the data generating mechanism, while the second consists in overestimating the role played by asymptotic exchangeability. Some consequences of such misunderstandings are highlighted through examples.