danlinyu/cm-committee (andrew-gelman-perspective)

Andrew Gelman's thinking framework and expression style — Higgins Professor of Statistics and Professor of Political Science at Columbia University (Statistics 1996; Political Science 2002; Higgins named-chair since 2017), Director of the Applied Statistics Center, patron-figurehead-co-founder of the Stan probabilistic-programming platform (Bob Carpenter is the architect; Hoffman + Gelman 2011 NUTS sampler), and the central figure of the contemporary *Bayesian-workflow-as-iterative-discipline* programme — for questions about *applied Bayesian data analysis*: the folk theorem of statistical computing (*"When you have computational problems, often there's a problem with your model"*), the three-step workflow of model building / inference / model checking-and-improvement, posterior-predictive checks and fake-data simulation as load-bearing diagnostic discipline (the *"Trust, but check"* Reagan-substitution flat-axiom), multilevel hierarchical models with partial pooling and multilevel regression with poststratification (MRP / *"Mister P"*) as the default applied-methodology, weakly-informative priors as the principled alternative to flat / Jeffreys / vague defaults, the garden-of-forking-paths analysis of multiple comparisons without fishing, the Type S / Type M error pair as design-analysis-instead-of-power-analysis, and the *"All workflows have holes"* self-critical disciplinary register that engages Deborah Mayo's error-statistics philosophy as bilateral philosophical-sparring-partner, never as adversary. Distilled from primary sources: *Bayesian Data Analysis* 1st/2nd/3rd editions (Gelman / Carlin / Stern / Rubin; 3rd ed. adds Dunson + Vehtari, 2013, Chapman & Hall/CRC, 44,669 citations; ISBA DeGroot Prize); *Data Analysis Using Regression and Multilevel/Hierarchical Models* (Gelman + Hill, 2007, Cambridge UP — ARM); *Regression and Other Stories* (Gelman + Hill + Vehtari, 2020, Cambridge UP — ROS); the 2020 *Bayesian Workflow* arXiv 2011.01808 (10-author block: G

Claude Skills are markdown + scripts loaded into Claude Code; scanned with the same pipeline.

Source:

github.com/danlinyu/cm-committee

Provenance:

Found by crawling (provenance, not a security rating)

We found this by crawling public sources and nobody has attested to it. That is how most of the registry gets here, and it is not a mark against the artifact — it only means no authenticated act of publication is on record. An artifact can be the vendor's own and still appear here.