I would like to highlight two resources that, in my humble opinion as ISBA Social Media Manager, remain under-recognized yet immensely valuable:
ISBA Webinars on Bayesian Analysis Articles (2019–present).This webpage gathers an exceptional collection of webinars discussing Bayesian Analysis articles since 2019. For anyone curious about the frontiers of Bayesian statistics, this series brings together cutting-edge research from world-class experts. The talks span topics such as model uncertainty and missing data, new perspectives on stick-breaking models, sparse Bayesian factor analysis, and advances in causal inference under model mis-specification. Other contributions cover nonparametric priors, spatio-temporal modeling of Arctic sea ice, Bayesian regression trees for causal inference, and much more.The next BA webinar will focus on the paper “Model Uncertainty and Missing Data: An Objective Bayesian Perspective” by G. García-Donato, M. Eugenia Castellanos, S. Cabras, A. Quirós, and A. Forte. There will be four invited discussants: M. Clyde, M. Ferreira, A. Ly, and J. Rubio. It is scheduled for November 5, 2025 (4:00 PM UTC | 11:00 AM EST | 5:00 PM CET). Registration will be announced later on this webpage.ISBA YouTube Channel.All recorded webinar videos are available on the ISBA YouTube channel. Beyond the webinars, the channel hosts curated playlists from ISBA World Meetings (2012, 2016, 2018, 2021, 2022, 2024), specialized workshops and seminars (ABI, BNP, BayesComp), as well as content from ISBA Sections (j-ISBA, BNP, BioPharma, Industrial).These resources deserve broad visibility. I warmly encourage you to explore them, share them within your networks, and let us know your feedback.— Julyan, on behalf of the ISBA Social Media team
Archive for discussion paper
Sparse Bayesian factor analysis when the number of factors Is unknown [ISBA webinar]
Posted in Books, Statistics, University life with tags Austria, Bayesian Analysis, discussion paper, factor analysis, International Society for Bayesian Analysis, shrinkage estimation, Sylvia Frühwirth-Schnatter, unknown number of components, Vienna, webinar, WU Wien on February 10, 2025 by xi'an
a lesser-known correlate of the Jeffreys-Lindley paradox (with discussion)
Posted in Books, pictures, Statistics, University life with tags Bayesian Analysis, Bayesian testing, confidence intervals, credible intervals, discussion paper, improper prior, Jeffreys-Lindley paradox, matching priors, paradoxes, point null hypotheses, two-sided hypotheses, UBC, University of British Columbia, Vancouver, webinar, youtube on October 19, 2024 by xi'an
Two UBC faculty, Harlan Campbell and Paul Gustafson, wrote a paper entitled “Defining a Credible Interval Is Not Always Possible with “Point-Null” Priors: A Lesser-Known Correlate of the Jeffreys-Lindley Paradox” in Bayesian Analysis (2024, 19, Number 3, pp. 925–984), which got discussed and presented on the BA webinar yesterday. I missed the call for discussion, on a topic I would have liked very much to discuss and an analysis I strongly disagree with. Fortunately, several of the discussants in the webinar and in the printed version advanced some of my points (as. e.g., Bertrand Clarke in the above slide screen-shot from the on-line video).
I find the paper somewhat missing in linking with the history of the topic, with no mention of Berger & Sellke (1987) that comes as a counterpoint to Casella &—the other—Berger (1987), opposing one sided to two sided tests. Or of matching priors, which connect credible and confidence intervals to higher orders. But the central issue with the apparent contradiction between rejecting the point null hypothesis and returning a credible interval that contains the null is that the construction proceeds from a model averaged posterior. Which fundamentally contradicts the construct of a pair of priors attached with each model towards selecting the fittest one. And requires a far-from-innocent choice of respective prior weights for both models, an ill-defined notion I have repeatedly criticised here and elsewhere. Model averaging clashes with model selection in both decision-theoretic and modelling terms. In model averaging terms, the disappearance of the opposition exhibited by the authors in the predictive distribution, as shown by discussants Held and Pawel, is unsurprising. And makes the spike-and-slab prior far of a necessity. Contrariwise to the model selection case where it proves unavoidable. And for which a merged credible interval does not make sense (to me at least) since it should be constructed once one (and only one) of the two models is chosen. At this point, that the other model ever was considered should not impact subsequent inference. And within that perspective I do not see the relevance of agnostic (ignoring the model choice ation) 5% confidence or credible regions.
“…considers the regime of a fixed true parameter value as n increases [and] of a fixed p-value…” (p928)
With regards with the connection with the Jeffreys-Lindley (or Lindley-Jeffreys) so-called paradox, on which I have already written a lot (or even too much!), many of the earlier objections resurface. Like the measure-theoretic difficulty in including within a continuous interval an atom, i.e., a value with a point mass. Which isolates this atom away from any other value in the interval (and of course creates discontinuities). Or fixing the p-value forever after (when n goes to infinity), as in the graph below (p929). Or treating an improper prior without further caution than with a proper prior. Especially when these are “created” by the decision problem itself.
Arnak Dalalyan at the RSS Journal Webinar
Posted in Books, pictures, Statistics, Travel, University life with tags approximate sampling, discussion paper, exact sampling, Journal of the Royal Statistical Society, Langevin MCMC algorithm, London, OUP, Oxford University Press, Palaiseau, Paris-Saclay campus, Read paper, Royal Statistical Society, RSS, RSS Journal Webinar, webinar on October 15, 2023 by xi'an
My friend and CREST colleague Arnak Dalalyan will (re)present [online] a Read Paper at the RSS on 31 October with my friends Hani Doss and Alain Durmus as discussants:
‘Theoretical Guarantees for Approximate Sampling and Log-Concave Densities’
Arnak Dalalyan ENSAE Paris, France
Sampling from various kinds of distributions is an issue of paramount importance in statistics since it is often the key ingredient for constructing estimators, test procedures or confidence intervals. In many situations, exact sampling from a given distribution is impossible or computationally expensive and, therefore, one needs to resort to approximate sampling strategies. However, there is no well-developed theory providing meaningful non-asymptotic guarantees for the approximate sampling procedures, especially in high dimensional problems. The paper makes some progress in this direction by considering the problem of sampling from a distribution having a smooth and log-concave density defined on ℝᵖ, for some integer p > 0. We establish non-asymptotic bounds for the error of approximating the target distribution by the distribution obtained by the Langevin Monte Carlo method and its variants. We illustrate the effectiveness of the established guarantees with various experiments. Underlying our analysis are insights from the theory of continuous time diffusion processes, which may be of interest beyond the framework of log-concave densities that are considered in the present work.
webscussion on Bayesian causality
Posted in Books, pictures, Statistics, University life with tags Bayesian Analysis, discussion paper, International Society for Bayesian Analysis, ISBA, webinar on June 18, 2023 by xi'anThree days ago, I attended (most of) and briefly took part in the discussion webinar run by Bayesian Analysis (the journal) and featuring the paper Causal Inference Under Mis-Specification: Adjustment Based on the Propensity Score by David Stephens, Widemberg Nobre, Erica Moodie, and Alexandra M. Schmidt, for which Pierre Jacob and I contributed a written discussion. I must admit being still rather agnostic about the possibility of running (Bayesian) causality in a general framework and in particular in the potentially misspecified setup studied by the authors. Among other things, there does not seem to be a framework where the suitability of the assumed balancing score can be tested or assessed, the Bayesian perspective is debatable, in particular because of the choice of plug-in estimates rather than cut/modularised versions, and the apparent inclusion of the regressors in the modelling is far from appealing (to me).
