https://agostinognasso.r-universe.devRecent package updates in agostinognassoR-universehttps://github.com/agostinognasso.pnghttps://agostinognasso.r-universe.devWed, 09 Sep 2026 09:31:33 GMTagostino.gnasso@unina.it (Agostino Gnasso)Variable importance rankings depend on the method, the random seed and the resample used to compute them. This package treats every source of importance as a judge expressing a ranking over the predictors, and synthesises those rankings into a Kemeny median ranking with ties. Uncertainty about the consensus is quantified through bootstrap rank confidence sets, top-k probabilities and clustering of disagreeing judges.https://github.com/r-universe/agostinognasso/actions/runs/39113155482Wed, 09 Sep 2026 09:31:33 GMTrankimp1.0.0successhttps://agostinognasso.r-universe.devhttps://github.com/agostinognasso/rankimprankimp-intro.Rmdrankimp-intro.htmlConsensus ranking of variable importance2026-07-09 21:44:362026-09-08 10:28:06reference.Rmdreference.htmlEvery function, and when you want it2026-09-08 10:28:062026-09-08 10:28:06stability.Rmdstability.htmlHow stable is a variable importance ranking?2026-07-09 21:44:362026-09-08 15:29:59credit-scoring.Rmdcredit-scoring.htmlRobustly important variables in credit scoring2026-07-09 21:44:362026-09-08 15:29:59against-set-stability.Rmdagainst-set-stability.htmlSet stability is not rank stability2026-07-10 07:41:232026-09-08 10:28:06theory.Rmdtheory.htmlThe theory the package rests on2026-09-08 10:28:062026-09-08 10:28:06method-disagreement.Rmdmethod-disagreement.htmlWhen XAI methods disagree2026-07-09 21:44:362026-09-08 10:28:06agostino.gnasso@unina.it (Agostino Gnasso)Treats the proximity matrices produced by tree ensembles as first-class statistical objects rather than as model by-products. Provides a unified extractor across ensemble engines, in-bag and out-of-bag definitions, transformations to dissimilarities with metric diagnostics, corrections that make an indefinite proximity usable as a kernel, and permutation inference for comparing two proximity matrices or partitioning one across the terms of a design. Also provides a Nystrom approximation, a thresholded sparse representation and a streaming form that never allocates the matrix at all, for samples too large to hold it, measures how far the proximity moves between replicates of the ensemble, and draws each object through 'ggplot2': the seriated matrix, the configuration it implies, and the thresholded graph with its communities.https://github.com/r-universe/agostinognasso/actions/runs/39113047542Wed, 09 Sep 2026 09:26:52 GMTProximum1.1.0successhttps://agostinognasso.r-universe.devhttps://github.com/agostinognasso/proximumcredit-scoring-case.Rmdcredit-scoring-case.htmlA credit scoring case study2026-07-09 21:44:182026-09-08 10:34:13comparing-forests.Rmdcomparing-forests.htmlComparing forests through their proximities2026-07-09 21:44:182026-09-08 10:34:13Proximum-intro.RmdProximum-intro.htmlProximity matrices as statistical objects2026-07-10 11:18:462026-09-08 10:34:13large-n.Rmdlarge-n.htmlProximity matrices when n is large2026-07-09 21:44:182026-09-08 15:29:36