{"id":20917535,"url":"https://github.com/nimble-dev/ahmnimble","last_synced_at":"2025-05-13T12:30:52.618Z","repository":{"id":81858967,"uuid":"138140129","full_name":"nimble-dev/AHMnimble","owner":"nimble-dev","description":"Examples from Kéry \u0026 Royle's \"Applied Hierarchical Modeling in Ecology (Volume I)\" converted to NIMBLE","archived":false,"fork":false,"pushed_at":"2018-06-21T08:51:10.000Z","size":34898,"stargazers_count":23,"open_issues_count":0,"forks_count":4,"subscribers_count":10,"default_branch":"master","last_synced_at":"2024-03-14T22:19:59.148Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"R","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/nimble-dev.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2018-06-21T08:12:38.000Z","updated_at":"2023-06-08T18:47:10.000Z","dependencies_parsed_at":"2023-10-20T23:15:07.498Z","dependency_job_id":null,"html_url":"https://github.com/nimble-dev/AHMnimble","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nimble-dev%2FAHMnimble","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nimble-dev%2FAHMnimble/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nimble-dev%2FAHMnimble/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nimble-dev%2FAHMnimble/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nimble-dev","download_url":"https://codeload.github.com/nimble-dev/AHMnimble/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":225208564,"owners_count":17438209,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-18T16:33:59.451Z","updated_at":"2024-11-18T16:33:59.534Z","avatar_url":"https://github.com/nimble-dev.png","language":"R","readme":"# AHMnimble\n\nRun WinBUGS/OpenBUGS/JAGS examples from Volume I of \"Applied Hierarchical\nModeling in Ecology\" (Kéry and Royle) in NIMBLE.\n\n[*Applied Hierarchical Modeling in Ecology: Analysis of distribution, abundance and species richness in R and BUGS.  Volume I: Prelude and Static Models*](https://www.elsevier.com/books/applied-hierarchical-modeling-in-ecology-analysis-of-distribution-abundance-and-species-richness-in-r-and-bugs/kery/978-0-12-801378-6)\nby Marc K\u0026#233;ry and J. Andrew Royle (2015, Academic Press) is a\npopular introduction to some common hierarchical models in ecology.  The [book's web site](https://www.mbr-pwrc.usgs.gov/pubanalysis/keryroylebook/) provides code for its examples.\n\nWe have adapted most of the examples that use the popular WinBUGS,\nOpenBUGS or JAGS packages for Markov chain Monte Carlo (MCMC) to\nNIMBLE.  This was easy to do because NIMBLE uses (nearly) the same\nmodeling language.  We\nthank the authors along with Mike Meridith for permission to reproduce\ntheir code with modifications for use with NIMBLE. \n\nNIMBLE also provides tools for comparing the performance of MCMC\nalgorithms.   We have generally provided such comparisons between JAGS\nand NIMBLE's default sampler configuration.  These were run on a\nMac with a 3.4 GHz Intel iCore 5 with OS X 10.13.4.\n\nWe have attempted to make each file labeled with\n\"\\*\\_example\\_nimble.R\" or \"\\*\\_custom\\_nimble.R\" be self-contained,\nsourcing from other files as needed. \n\nSome examples remain to be converted.  Please do so and contribute to\nthis repository.\n\n## Potential to customize NIMBLE\nIn many cases, it is possible to improve NIMBLE's performance by\nre-writing the model using NIMBLE's extensibility and/or customizing\nthe configuration of MCMC samplers.  As of the initial launch of this\nrepository, we have only included a few customizations in the MCMC\ncomparisons.  Please try your own and file a pull request to include\nthem, or contact us at nimble.stats@gmail.com.\n\nExamples of customizing NIMBLE for Hidden Markov Models (HMMs)\nembedded in multi-state capture-recapture models are given in\n[ D. Turek, P. de Valpine, and C. J. Paciorek (2016). *Efficient Markov chain Monte Carlo sampling for hierarchical hidden Markov models*.  Environmental and Ecological Statistics 23:549-564.](https://link.springer.com/article/10.1007/s10651-016-0353-z)\n\n## How do we compare MCMCs?\n\nWe compare MCMCs based on *MCMC efficiency*, which we define as\n\n- MCMC efficiency = Effective sample size / computation time\n\nThis carries the interpretation of the number of effectively\nindependent samples generated per second.\n\nMCMC efficiency is different for each parameter because the effective\nsample size is different for each parameter.  For a single metric of\nMCMC performance for an entire model, we use the **minimum MCMC\nefficiency** across all monitored parameters.  The minimum is\nimportant because one needs to be sure that all parameters have mixed\nwell before trusting MCMC results. \n\nNote that we generally do not thin samples when comparing MCMC\nefficiency.  One could debate this choice, but we make it because\nthinning always results in some loss of statistical information.  In\npractice, one often thins.  When comparing MCMC methods, some mix more\nslowly but do so computationally faster, and vice-versa.  We attempt a\nclean comparison of MCMC efficiency by not thinning.  For some models,\nwe make an exception and use some thinning to avoid excessively large\nMCMC sample sizes.\n\n## Please contribute\n\nPlease file pull requests or email us at nimble.stats@gmail.com if you\nexplore some of these models and have code and/or results that would\nbe of interest to the community.\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnimble-dev%2Fahmnimble","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnimble-dev%2Fahmnimble","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnimble-dev%2Fahmnimble/lists"}