{ "_id": "6ad26942475807f7554a61b5", "Package": "DeCovarT", "Title": "Covariance-Aware Deconvolution of Bulk Transcriptomic Samples", "Version": "2.3.1", "Authors@R": "person(\"Bastien\", \"Chassagnol\", , \"bastien_chassagnol@laposte.net\", role = c(\"aut\", \"cre\", \"cph\"),\ncomment = c(ORCID = \"0000-0002-8955-2391\"))", "Description": "Estimates cell-type proportions in bulk transcriptomic\nsamples with a probabilistic convolution model that integrates\nthe gene-gene covariance structure of purified reference\nprofiles. Cellular ratios are recovered by maximum likelihood\nunder a multivariate Gaussian convolution, using analytic\ngradients and Hessians, an isometric log-ratio\nreparametrisation (Helmert basis) that enforces the simplex\nconstraint, and Marquardt-Levenberg or Newton-type optimisers.\nThe methodology is described in Chassagnol, Nuel and Becht\n(2023) .", "License": "MIT + file LICENSE", "URL": "https://github.com/bastienchassagnol/DeCovarT,\nhttps://bastienchassagnol.github.io/DeCovarT/", "BugReports": "https://github.com/bastienchassagnol/DeCovarT/issues", "VignetteBuilder": "knitr, quarto", "RdMacros": "Rdpack", "Config/roxygen2/version": "8.1.0", "Config/testthat/edition": "3", "Encoding": "UTF-8", "Language": "en-GB", "Roxygen": "list(markdown = TRUE, roclets = c(\"namespace\", \"rd\",\n\"srr::srr_stats_roclet\"))", "RoxygenNote": "8.1.0", "Config/pak/sysreqs": "libicu-dev", "Repository": "https://bastienchassagnol.r-universe.dev", "Date/Publication": "2026-10-16 15:12:25 UTC", "RemoteUrl": "https://github.com/bastienchassagnol/DeCovarT", "RemoteRef": "HEAD", "RemoteSha": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "NeedsCompilation": "no", "Packaged": { "Date": "2026-10-16 17:44:40 UTC", "User": "root" }, "Author": "Bastien Chassagnol [aut, cre, cph] (ORCID:\n)", "Maintainer": "Bastien Chassagnol ", "_user": "bastienchassagnol", "_type": "src", "_file": "DeCovarT_2.3.1.tar.gz", "_fileid": "https://r2.ropensci.org/29ef97c1fe5839f37b84f4eb123d75479e8506aa64bf93917b356a2c2171636f", "_filesize": 13336109, "_sha256": "29ef97c1fe5839f37b84f4eb123d75479e8506aa64bf93917b356a2c2171636f", "_expires": "2027-01-24T18:13:19.000Z", "_created": "2026-10-16T17:44:40.000Z", "_published": "2026-10-16T18:13:22.439Z", "_buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467", "_status": "success", "_host": "GitHub-Actions", "_jobs": [ { "job": 117112010790, "time": 237, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "WARNING", "artifact": "12063398284" }, { "job": 117112010603, "time": 217, "config": "linux-release-x86_64", "r": "4.6.1", "check": "WARNING", "artifact": "12063502906" }, { "job": 117112010584, "time": 171, "config": "macos-oldrel-arm64", "r": "4.5.3", "check": "WARNING", "artifact": "12065840107" }, { "job": 117112010549, "time": 182, "config": "macos-release-arm64", "r": "4.6.1", "check": "WARNING", "artifact": "12064104908" }, { "job": 117110007527, "time": 272, "config": "source", "r": "4.6.1", "check": "OK", "artifact": "12063492348" }, { "job": 117112010526, "time": 151, "config": "wasm-release", "r": "4.6.0", "check": "OK", "artifact": "12062948677" }, { "job": 117112010601, "time": 139, "config": "windows-devel-x86_64", "r": "4.7.0", "check": "WARNING", "artifact": "12063333687" }, { "job": 117112010759, "time": 132, "config": "windows-oldrel-x86_64", "r": "4.5.3", "check": "WARNING", "artifact": "12063727509" }, { "job": 117112010541, "time": 193, "config": "windows-release-x86_64", "r": "4.6.1", "check": "WARNING", "artifact": "12064205639" } ], "_upstream": "https://github.com/bastienchassagnol/DeCovarT", "_commit": { "id": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "author": "bastienchassagnol ", "committer": "bastienchassagnol ", "message": "Update references and enhance article layout for clarity and completeness\n\n- Revised the `appendix_layout.tex` to replace the Erdős–Rényi model description with a generalised random graph model, citing Newman (2002) for clarity.\n- Added new references in `decovart_library.bib` and `REFERENCES.bib`, including works on generalized linear models and robust estimation, enriching the theoretical foundation of the package.\n- Updated the `main_layout.tex` to include a new author, Elias Ventre, and acknowledged his contributions to the manuscript.\n- Enhanced the discussion on M-estimation in the `main_layout.tex` by citing Hampel (1964) and Maronna et al. (2018) for robust statistical methods.\n- Documented all changes in `NEWS.md` to keep users informed of the latest updates in the DeCovarT package.\n", "time": 1792163545 }, "_maintainer": { "name": "Bastien Chassagnol", "email": "bastien_chassagnol@laposte.net", "login": "bastienchassagnol", "orcid": "0000-0002-8955-2391", "description": "", "uuid": 23634475 }, "_distro": "resolute", "_registered": true, "_dependencies": [ { "package": "R", "version": ">= 4.1.0", "role": "Depends" }, { "package": "dplyr", "role": "Imports" }, { "package": "ggplot2", "version": ">= 3.5.0", "role": "Imports" }, { "package": "marqLevAlg", "role": "Imports" }, { "package": "MASS", "role": "Imports" }, { "package": "Matrix", "role": "Imports" }, { "package": "purrr", "role": "Imports" }, { "package": "Rdpack", "version": ">= 0.7", "role": "Imports" }, { "package": "rlang", "role": "Imports" }, { "package": "tensor", "role": "Imports" }, { "package": "tibble", "role": "Imports" }, { "package": "tidyr", "role": "Imports" }, { "package": "circlize", "role": "Suggests" }, { "package": "cli", "role": "Suggests" }, { "package": "ComplexHeatmap", "role": "Suggests" }, { "package": "compositions", "role": "Suggests" }, { "package": "cowplot", "role": "Suggests" }, { "package": "e1071", "role": "Suggests" }, { "package": "EMMIXmfa", "role": "Suggests" }, { "package": "flextable", "role": "Suggests" }, { "package": "forcats", "role": "Suggests" }, { "package": "funkyheatmap", "role": "Suggests" }, { "package": "furrr", "role": "Suggests" }, { "package": "future", "role": "Suggests" }, { "package": "ggdendro", "role": "Suggests" }, { "package": "ggdist", "role": "Suggests" }, { "package": "ggtext", "role": "Suggests" }, { "package": "glmnet", "role": "Suggests" }, { "package": "gridExtra", "role": "Suggests" }, { "package": "htmlwidgets", "role": "Suggests" }, { "package": "igraph", "role": "Suggests" }, { "package": "knitr", "role": "Suggests" }, { "package": "limSolve", "role": "Suggests" }, { "package": "litedown", "role": "Suggests" }, { "package": "mclust", "role": "Suggests" }, { "package": "Metrics", "role": "Suggests" }, { "package": "MixSim", "role": "Suggests" }, { "package": "mvtnorm", "role": "Suggests" }, { "package": "nnls", "role": "Suggests" }, { "package": "numDeriv", "role": "Suggests" }, { "package": "patchwork", "role": "Suggests" }, { "package": "pkgdown", "role": "Suggests" }, { "package": "plotly", "role": "Suggests" }, { "package": "png", "role": "Suggests" }, { "package": "ps", "role": "Suggests" }, { "package": "qrng", "role": "Suggests" }, { "package": "quarto", "role": "Suggests" }, { "package": "qqplotr", "role": "Suggests" }, { "package": "reactable", "version": ">= 0.4.4", "role": "Suggests" }, { "package": "readr", "role": "Suggests" }, { "package": "rgl", "role": "Suggests" }, { "package": "rmarkdown", "role": "Suggests" }, { "package": "spelling", "role": "Suggests" }, { "package": "stringr", "role": "Suggests" }, { "package": "testthat", "version": ">= 3.0.0", "role": "Suggests" }, { "package": "tinytable", "version": ">= 0.15.0", "role": "Suggests" }, { "package": "viridis", "role": "Suggests" }, { "package": "withr", "role": "Suggests" } ], "_owner": "bastienchassagnol", "_selfowned": true, "_usedby": 0, "_updates": [ { "week": "2026-04", "n": 1 }, { "week": "2026-15", "n": 3 }, { "week": "2026-16", "n": 2 }, { "week": "2026-17", "n": 8 }, { "week": "2026-20", "n": 3 }, { "week": "2026-29", "n": 17 }, { "week": "2026-30", "n": 9 }, { "week": "2026-31", "n": 6 }, { "week": "2026-32", "n": 57 }, { "week": "2026-33", "n": 4 }, { "week": "2026-34", "n": 15 }, { "week": "2026-35", "n": 9 }, { "week": "2026-36", "n": 17 }, { "week": "2026-37", "n": 15 }, { "week": "2026-38", "n": 6 }, { "week": "2026-39", "n": 5 }, { "week": "2026-40", "n": 2 }, { "week": "2026-42", "n": 8 } ], "_tags": [ { "name": "v0.1.1", "date": "2026-08-03" }, { "name": "v0.1.2", "date": "2026-08-03" }, { "name": "v0.1.3", "date": "2026-08-05" }, { "name": "v1.0.0", "date": "2026-08-07" }, { "name": "v2.0.0", "date": "2026-08-07" }, { "name": "v2.0.1", "date": "2026-08-07" }, { "name": "v2.0.2", "date": "2026-08-12" }, { "name": "v2.1.0", "date": "2026-08-15" }, { "name": "v2.2.0", "date": "2026-08-19" }, { "name": "v2.2.1", "date": "2026-08-20" } ], "_stars": 1, "_contributors": [ { "user": "bastienchassagnol", "count": 58, "uuid": 23634475 }, { "user": "bastienchassagnol-servier", "count": 2, "uuid": 82661205 }, { "user": "copilot", "count": 2, "uuid": 198982749 } ], "_userbio": { "uuid": 23634475, "type": "user", "name": "chassagnol", "followers": 17 }, "_downloads": { "count": 0, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/DeCovarT" }, "_devurl": "https://github.com/bastienchassagnol/decovart", "_pkgdown": "https://bastienchassagnol.github.io/DeCovarT/", "_searchresults": 21, "_quarto": true, "_topics": [ "quarto" ], "_rbuild": "4.6.1", "_assets": [ "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/contents.json", "extra/DeCovarT.html", "extra/manual.pdf", "extra/NEWS.html", "extra/NEWS.txt", "extra/readme.html", "extra/readme.md", "LICENSE" ], "_cranurl": false, "_exports": [ ".compute_global_variance", ".inner_product", ".jeffreys_gaussian", ".sigma_p_factorisation", "additive_log_ratio", "additive_logistic", "algorithm_similarity", "assign_iid_signed_weights", "bootstrap_decovart", "boundary_diagnostics", "build_covariance_array_from_precision", "build_normalised_precision", "check_true_theta", "chi_bar_square_pvalue", "composition_from_entropy", "compute_average_jeffreys", "compute_average_overlap", "compute_average_riemannian", "compute_benchmark_metrics", "compute_glmnet_gene_scores", "compute_mean_profile_objectives", "compute_shannon_entropy", "confint_profile_decovart", "covariance_structure_from_graph_model", "coverage_mc_interval", "deconvolute_ratios", "deconvolute_ratios_cibersort", "deconvolute_ratios_deconrnaseq", "deconvolute_ratios_gls", "deconvolute_ratios_gradient_descent", "deconvolute_ratios_L_BFGS_B", "deconvolute_ratios_Marquardt_Levenberg", "deconvolute_ratios_Newton_Raphson", "deconvolute_ratios_nnls", "deconvolute_ratios_rlm", "deconvolute_ratios_simulated_annealing", "describe_simulation_scenario", "equicorrelation_gram", "equivariance_check_decovart", "expected_fisher_unconstrained", "fit_decovart", "fixed_gls_covariance", "generate_mean_signature_matrix", "generate_random_network_skeleton", "gradient_loglik_constrained", "gradient_loglik_unconstrained", "helmert_basis", "hessian_additive_logistic", "hessian_isometric_logistic", "hessian_loglik_constrained", "hessian_loglik_unconstrained", "isometric_log_ratio", "isometric_logistic", "jacobian_additive_logistic", "jacobian_isometric_logistic", "loglik_multivariate", "loglik_multivariate_constrained", "lrt_decovart", "multistart_decovart", "new_decovart_covariance", "overlap_gaussian_mc", "pivot_mc_estimates", "plot_algorithm_similarity", "plot_correlation_Heatmap", "plot_mc_forest", "plot_mc_metric_dots", "plot_mc_raincloud", "profile_loglik_decovart", "reference_bootstrap_decovart", "repair_simplex", "restricted_mle_decovart", "run_simulation_benchmark", "scale_covariance_array", "sigma_logdet", "sigma_quadform", "sigma_solve", "sigma_trace_precision_times", "simulate_bulk_mixture", "simulate_hierarchical_grn_moments", "spd_affine_invariant_distance", "starting_simplex", "theme_decovart_facets", "vcov_ilr_delta" ], "_help": [ { "page": "additive_logistic", "title": "Additive logistic transform (unconstrained coordinates to the simplex)", "topics": [ "additive_logistic", "additive_log_ratio" ] }, { "page": "algorithm_similarity", "title": "Algorithm-similarity correlation from a Monte Carlo benchmark", "topics": [ "algorithm_similarity" ] }, { "page": "bootstrap_decovart", "title": "Parametric bootstrap for DeCovarT proportions and boundary tests", "concept": [ "decovart_inference" ], "topics": [ "bootstrap_decovart" ] }, { "page": "check_true_theta", "title": "Validate generative-model parameters theta", "topics": [ "check_true_theta" ] }, { "page": "chi_bar_square_pvalue", "title": "Chi-bar-square tail probability for boundary likelihood-ratio tests", "concept": [ "decovart_inference" ], "topics": [ "chi_bar_square_pvalue" ] }, { "page": "composition_from_entropy", "title": "One-dominant composition with a target normalised Shannon entropy", "topics": [ "composition_from_entropy" ] }, { "page": "compute_average_jeffreys", "title": "Average pairwise Jeffreys divergence of a Gaussian mixture", "topics": [ "compute_average_jeffreys" ] }, { "page": "compute_average_overlap", "title": "Average pairwise overlap of a Gaussian mixture", "topics": [ "compute_average_overlap" ] }, { "page": "compute_average_riemannian", "title": "Mean pairwise AIRM distance of component covariances", "topics": [ "compute_average_riemannian" ] }, { "page": "compute_benchmark_metrics", "title": "Compute deconvolution benchmark metrics", "topics": [ "compute_benchmark_metrics" ] }, { "page": "compute_glmnet_gene_scores", "title": "Gene scores from multinomial elastic-net cell-type classification", "topics": [ "compute_glmnet_gene_scores" ] }, { "page": "compute_shannon_entropy", "title": "Normalised Shannon entropy of a discrete distribution", "topics": [ "compute_shannon_entropy" ] }, { "page": "confint_profile_decovart", "title": "Profile-likelihood confidence intervals for cellular ratios", "concept": [ "decovart_inference" ], "topics": [ "confint_profile_decovart" ] }, { "page": "coverage_mc_interval", "title": "Monte Carlo interval for an empirical coverage probability", "topics": [ "coverage_mc_interval" ] }, { "page": "deconvolute_ratios", "title": "Parallel deconvolution of a bulk expression matrix", "topics": [ "deconvolute_ratios" ] }, { "page": "deconvolute_ratios_Marquardt_Levenberg", "title": "DeCovarT MLE of cellular proportions for one bulk sample", "topics": [ "deconvolute_ratios_cibersort", "deconvolute_ratios_deconrnaseq", "deconvolute_ratios_gradient_descent", "deconvolute_ratios_L_BFGS_B", "deconvolute_ratios_Marquardt_Levenberg", "deconvolute_ratios_Newton_Raphson", "deconvolute_ratios_nnls", "deconvolute_ratios_rlm", "deconvolute_ratios_simulated_annealing" ] }, { "page": "deconvolute_ratios_gls", "title": "Generalised least squares with a fixed residual covariance", "topics": [ "deconvolute_ratios_gls" ] }, { "page": "describe_simulation_scenario", "title": "Describe a Gaussian-convolution simulation scenario", "topics": [ "describe_simulation_scenario" ] }, { "page": "equicorrelation_gram", "title": "Equicorrelation (constant-correlation) Gram matrix", "topics": [ "equicorrelation_gram" ] }, { "page": "expected_fisher_unconstrained", "title": "Expected Fisher information of unconstrained \\boldsymbol{p}", "topics": [ "expected_fisher_unconstrained" ] }, { "page": "fit_decovart", "title": "Fit the DeCovarT Gaussian-convolution model", "concept": [ "decovart_fit" ], "topics": [ "coef.decovart_fit", "confint.decovart_fit", "fitted.decovart_fit", "fit_decovart", "nobs.decovart_fit", "plot.decovart_fit", "print.decovart_fit", "print.summary.decovart_fit", "residuals.decovart_fit", "summary.decovart_fit", "vcov.decovart_fit" ] }, { "page": "fixed_gls_covariance", "title": "Fixed residual covariance for a GLS competitor", "topics": [ "fixed_gls_covariance" ] }, { "page": "generate_mean_signature_matrix", "title": "Generate mean profiles from a target Gram matrix", "topics": [ "generate_mean_signature_matrix" ] }, { "page": "helmert_basis", "title": "Helmert contrast matrix for isometric log-ratio coordinates", "topics": [ "helmert_basis" ] }, { "page": "hessian_isometric_logistic", "title": "Hessian tensor of the isometric logistic map", "topics": [ "hessian_isometric_logistic" ] }, { "page": "isometric_logistic", "title": "Isometric logistic transform (ILR coordinates to the simplex)", "topics": [ "isometric_logistic", "isometric_log_ratio" ] }, { "page": "jacobian_isometric_logistic", "title": "Jacobian of the isometric logistic map", "topics": [ "jacobian_isometric_logistic" ] }, { "page": "lrt_decovart", "title": "Likelihood-ratio test for cellular ratios", "concept": [ "decovart_inference" ], "topics": [ "lrt_decovart" ] }, { "page": "multistart_decovart", "title": "Refit DeCovarT from several random starts to probe multimodality", "concept": [ "decovart_inference" ], "topics": [ "multistart_decovart" ] }, { "page": "overlap_gaussian_mc", "title": "MixSim-style pairwise overlap via stratified Sobol Monte Carlo", "topics": [ "overlap_gaussian_mc" ] }, { "page": "pivot_mc_estimates", "title": "Pivot Monte Carlo proportion estimates to a long table", "topics": [ "pivot_mc_estimates" ] }, { "page": "plot_algorithm_similarity", "title": "Tile heatmap of algorithm-similarity correlations", "topics": [ "plot_algorithm_similarity" ] }, { "page": "plot_correlation_Heatmap", "title": "Plot deconvolution metric heatmaps", "topics": [ "plot_correlation_Heatmap" ] }, { "page": "plot_mc_forest", "title": "Forest plot of ADEMP Monte Carlo summaries", "topics": [ "plot_mc_forest" ] }, { "page": "plot_mc_metric_dots", "title": "Faceted dot plot of several ADEMP metrics", "topics": [ "plot_mc_metric_dots" ] }, { "page": "plot_mc_raincloud", "title": "Horizontal raincloud of Monte Carlo proportion estimates", "topics": [ "plot_mc_raincloud" ] }, { "page": "reference_bootstrap_decovart", "title": "Reference-based bootstrap for signature and composition uncertainty", "concept": [ "decovart_inference" ], "topics": [ "reference_bootstrap_decovart" ] }, { "page": "repair_simplex", "title": "Repair a numeric vector onto the unit simplex", "topics": [ "repair_simplex" ] }, { "page": "run_simulation_benchmark", "title": "Simulate bulk mixtures and benchmark deconvolution algorithms", "topics": [ "run_simulation_benchmark" ] }, { "page": "scale_covariance_array", "title": "Scale component covariances, keeping precision zeros", "topics": [ "scale_covariance_array" ] }, { "page": "simulate_bulk_mixture", "title": "Simulate bulk mixtures from a multivariate Gaussian convolution", "topics": [ "simulate_bulk_mixture" ] }, { "page": "simulate_hierarchical_grn_moments", "title": "Simulate GRN first- and second-order moments", "topics": [ "simulate_hierarchical_grn_moments" ] }, { "page": "spd_affine_invariant_distance", "title": "Affine-invariant Riemannian (AIRM) distance between two SPD matrices", "topics": [ "spd_affine_invariant_distance" ] }, { "page": "starting_simplex", "title": "Open-simplex start for ILR solvers", "topics": [ "starting_simplex" ] }, { "page": "theme_decovart_facets", "title": "Faceted ggplot2 theme (black strips, panel border)", "topics": [ "theme_decovart_facets" ] }, { "page": "vcov_ilr_delta", "title": "Cramer-Rao / ILR delta-method covariance of \\hat{\\boldsymbol{p}}", "topics": [ "vcov_ilr_delta" ] } ], "_pkglogo": "https://github.com/bastienchassagnol/DeCovarT/raw/HEAD/man/figures/logo.png", "_readme": "https://github.com/bastienchassagnol/DeCovarT/raw/HEAD/README.md", "_rundeps": [ "cli", "codetools", "cpp11", "doParallel", "dplyr", "farver", "foreach", "generics", "ggplot2", "glue", "gtable", "isoband", "iterators", "labeling", "lattice", "lifecycle", "magrittr", "marqLevAlg", "MASS", "Matrix", "pillar", "pkgconfig", "purrr", "R6", "rbibutils", "RColorBrewer", "Rdpack", "rlang", "S7", "scales", "stringi", "stringr", "tensor", "tibble", "tidyr", "tidyselect", "utf8", "vctrs", "viridisLite", "withr" ], "_vignettes": [ { "source": "fig02-bivariate-toy.qmd", "filename": "fig02-bivariate-toy.html", "title": "§2.1 Bivariate toy model", "engine": "quarto::html", "headings": [ "Generative model", "Factorial design (972 scenarios)", "Scenario ID", "Log-likelihood along the simplex", "Inference", "Visualisations", "Other one-sample log-likelihood slices (not drawn)", "Likelihood geometry and solver behaviour", "What is being optimised", "One bulk, $N=1$", "Reading the twelve meta-scenarios", "What the heatmaps still add", "See also", "References" ], "created": "2026-09-04 10:34:11", "modified": "2026-10-15 08:01:47", "commits": 11 }, { "source": "theory-decovart-generative-model.qmd", "filename": "theory-decovart-generative-model.html", "title": "Derivatives of the DeCovarT generative model under simplex transforms", "engine": "quarto::html", "headings": [ "Notation", "Matrix calculus reminders", "First-order matrix calculus", "$$\\begin{aligned}\\frac{\\partial\\det(\\boldsymbol{A})}{\\partial p}&=\\det(\\boldsymbol{A}),\\mathrm{tr}!\\left(\\boldsymbol{A}^{-1}\\frac{\\partial\\boldsymbol{A}}{\\partial p}\\right),\\\\frac{\\partial(\\boldsymbol{U}\\boldsymbol{A}\\boldsymbol{V})}{\\partial p}&=\\boldsymbol{U}\\frac{\\partial\\boldsymbol{A}}{\\partial p}\\boldsymbol{V},\\\\frac{\\partial\\boldsymbol{A}^{-1}}{\\partial p}&=", "$$\\frac", "\\mathrm", "$$\\frac{\\partial}{\\partial p}\\bigl((\\boldsymbol{y}-\\boldsymbol{x}p)^{\\top}\\boldsymbol{\\Omega}(\\boldsymbol{y}-\\boldsymbol{x}p)\\bigr)", "Second-order matrix calculus", "$$\\frac", "\\boldsymbol", "$$\\frac", "Inner products in the score", "Unconstrained score and Hessian", "$$\\ell_{\\boldsymbol{y}\\mid\\boldsymbol{\\zeta}}(\\boldsymbol{p})", "\\tfrac{1}{2}\\log\\det\\boldsymbol{\\Omega}(\\boldsymbol{p})", "The determinant term carries a factor $\\tfrac{1}{2}$", "What “unconstrained” means here", "Unconstrained gradient", "Unconstrained Hessian", "$$\\begin", "\\underbrace", "\\boldsymbol", "$$\\begin", "\\underbrace", "\\boldsymbol", "Constrained optimisation on the simplex", "Log-sum-exp and softmax", "$$s_", "\\frac", "e^{z_{i}-\\operatorname{LSE}(\\boldsymbol{z})},\\qquad\\operatorname{LSE}(\\boldsymbol{z})", "$$H_", "s_{i}\\bigl[(\\delta_{ij}-s_{j})(\\delta_{ik}-s_{k})", "Isometric log-ratio map", "$$\\boldsymbol", "\\mathbf", "\\mathbf", "\\operatorname{softmax}(\\mathbf{V}\\boldsymbol{\\rho})", "ILR is a $C^{2}$ diffeomorphism", "Why ILR is the default chart", "Efficient derivatives of $\\boldsymbol", "$$\\mathbf{J}_{\\boldsymbol{\\psi}}(\\boldsymbol{\\rho})", "\\frac", "$$\\mathbf", "Change of variables: ambient $\\boldsymbol", "Constrained score and Hessian", "$$\\nabla_{\\boldsymbol{\\rho}}\\ell", "\\mathbf{J}{\\boldsymbol{\\psi}}^{\\mathsf{T}}\\nabla{\\boldsymbol{p}}\\ell", "$$\\mathbf", "Implementation map", "Expected Fisher information and Wald inference", "Unconstrained expected Fisher information", "$$I(\\boldsymbol", "Observed versus expected information", "Constrained case: ILR delta method", "$$I_", "\\mathbf", "Wald is the cheapest, not the safest, interval", "Numerical speed-ups and solver safeguards", "Cache a Cholesky factorisation of $\\boldsymbol{\\Sigma}(\\boldsymbol{p})$", "Guard the box-constrained L-BFGS-B path", "marqLevAlg Hessian sign under minimize = FALSE", "Model objects and S3 accessors", "Numerical consistency of the likelihood and its derivatives", "Density against mvtnorm", "Richardson extrapolation of the score and Hessian", "Expected information versus Monte Carlo (\\mathbb{E}[-\\mathbf{H}])", "Structure-aware covariance backends", "Decision tree", "Computational complexity", "Appendix: additive log-ratio chart", "$$p_", "\\frac", "\\frac", "Why ALR is not the default", "Structural zeros and folded simplex models", "References" ], "created": "2026-09-04 10:34:11", "modified": "2026-10-15 12:38:55", "commits": 7 } ], "_score": 4.878521795501206, "_indexed": true, "_nocasepkg": "decovart", "_universes": [ "bastienchassagnol" ], "_binaries": [ { "r": "4.7.0", "os": "linux", "version": "2.3.1", "date": "2026-10-16T17:48:33.000Z", "distro": "resolute", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/b91d6ed3e423960404737da0ca3ffb135cb6afa9991fc49218c83a1db29e2a50", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.6.1", "os": "linux", "version": "2.3.1", "date": "2026-10-16T17:48:10.000Z", "distro": "resolute", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/d903ed3a1542c5340907496d0070e37cf0fffeef8008e660c539db6f7d39fa03", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.5.3", "os": "mac", "version": "2.3.1", "date": "2026-10-16T18:12:09.000Z", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/c320333caf0ec816804452b868dcdcbe0588c7a108eb37c1a13889e41ba16292", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.6.1", "os": "mac", "version": "2.3.1", "date": "2026-10-16T18:08:32.000Z", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/1fd6301f73910862db7fa8e57648a580335dd958313150864ae159fcd82d43d4", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.6.0", "os": "wasm", "version": "2.3.1", "date": "2026-10-16T17:48:16.000Z", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/40044af334ada9067ca1ad604395b4033bef8b77f4c7f591ea32470c47e01ee7", "status": "success", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.7.0", "os": "win", "version": "2.3.1", "date": "2026-10-16T17:47:03.000Z", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/ec36d2d7fc4be5e61979ff1d0f9b313d050e888d0d59a7e3e1c8b3bf1abfa50c", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.5.3", "os": "win", "version": "2.3.1", "date": "2026-10-16T17:47:10.000Z", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/cf1337e6a517affcbaa7624fc71de7cdf19fc1d100210e9a0e373d7ad4bc7288", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" }, { "r": "4.6.1", "os": "win", "version": "2.3.1", "date": "2026-10-16T17:47:22.000Z", "commit": "1f2f76d8e56da034b6b5ce6bbd430c49b6204777", "fileid": "https://r2.ropensci.org/46fee3ed7b956ecc83622b654c2ca18ab6f29a6c6dd21ada460974cde2eea29f", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/bastienchassagnol/actions/runs/38941456467" } ] }