{ "_id": "6a1412edacfb0bcc41d3bbbc", "Package": "ADLP", "Type": "Package", "Title": "Accident and Development Period Adjusted Linear Pools for\nActuarial Stochastic Reserving", "Version": "0.1.0", "Authors@R": "c(\nperson(\"Benjamin\", \"Avanzi\", role = \"aut\", email = \"b.avanzi@unimelb.edu.au\"),\nperson(\"William\", \"Ho\", role = \"aut\", email = \"yunwaiho@gmail.com\"),\nperson(\"Yanfeng\", \"Li\", role = c(\"aut\", \"cre\"), email = \"yanfeng.li@student.unsw.edu.au\"),\nperson(\"Bernard\", \"Wong\", role = \"aut\", email = \"bernard.wong@unsw.edu.au\"),\nperson(\"Alan\", \"Xian\", role = \"aut\", email = \"alanxian@hotmail.com\"))", "Author": "Benjamin Avanzi [aut], William Ho [aut], Yanfeng Li [aut, cre],\nBernard Wong [aut], Alan Xian [aut]", "Maintainer": "Yanfeng Li ", "Description": "Loss reserving generally focuses on identifying a single\nmodel that can generate superior predictive performance.\nHowever, different loss reserving models specialise in\ncapturing different aspects of loss data. This is recognised in\npractice in the sense that results from different models are\noften considered, and sometimes combined. For instance,\nactuaries may take a weighted average of the prediction\noutcomes from various loss reserving models, often based on\nsubjective assessments. This package allows for the use of a\nsystematic framework to objectively combine (i.e. ensemble)\nmultiple stochastic loss reserving models such that the\nstrengths offered by different models can be utilised\neffectively. Our framework is developed in Avanzi et al.\n(2023). Firstly, our criteria model combination considers the\nfull distributional properties of the ensemble and not just the\ncentral estimate - which is of particular importance in the\nreserving context. Secondly, our framework is that it is\ntailored for the features inherent to reserving data. These\ninclude, for instance, accident, development, calendar, and\nclaim maturity effects. Crucially, the relative importance and\nscarcity of data across accident periods renders the problem\ndistinct from the traditional ensemble techniques in\nstatistical learning. Our framework is illustrated with a\ncomplex synthetic dataset. In the results, the optimised\nensemble outperforms both (i) traditional model selection\nstrategies, and (ii) an equally weighted ensemble. In\nparticular, the improvement occurs not only with central\nestimates but also relevant quantiles, such as the 75th\npercentile of reserves (typically of interest to both insurers\nand regulators). Reference: Avanzi B, Li Y, Wong B, Xian A\n(2023) \"Ensemble distributional forecasting for insurance loss\nreserving\" .", "License": "GPL-3", "Encoding": "UTF-8", "URL": "https://github.com/agi-lab/ADLP", "BugReports": "https://github.com/agi-lab/ADLP/issues", "LazyData": "true", "RoxygenNote": "7.3.1", "VignetteBuilder": "knitr", "NeedsCompilation": "no", "Packaged": { "Date": "2026-05-25 09:11:28 UTC", "User": "root" }, "Repository": "https://cran.r-universe.dev", "Date/Publication": "2024-04-19 02:34:28 UTC", "RemoteUrl": "https://github.com/cran/ADLP", "RemoteRef": "HEAD", "RemoteSha": "b09fd60db49dfadb92810318d6710e785432adef", "MD5sum": "754465e63a6aa1ac6b9a31c9d5feb706", "_user": "cran", "_type": "src", "_file": "ADLP_0.1.0.tar.gz", "_fileid": "de6836afa8eed7567d47c3adfbd44269c29a8594538a6756739bacc4e1279e18", "_filesize": 430100, "_sha256": "de6836afa8eed7567d47c3adfbd44269c29a8594538a6756739bacc4e1279e18", "_created": "2026-05-25T09:11:28.000Z", "_published": "2026-05-25T09:14:21.206Z", "_distro": "noble", "_jobs": [ { "job": 77686349952, "time": 130, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "OK", "artifact": "7195261450" }, { "job": 77686349905, "time": 140, "config": "linux-release-x86_64", "r": "4.6.0", "check": "OK", "artifact": "7195264049" }, { "job": 77685992148, "time": 189, "config": "source", "r": "4.6.0", "check": "OK", "artifact": "7195229851" }, { "job": 77686349904, "time": 110, "config": "wasm-release", "r": "4.6.0", "check": "OK", "artifact": "7195256372" } ], "_buildurl": "https://github.com/r-universe/cran/actions/runs/26392693516", "_status": "success", "_host": "GitHub-Actions", "_upstream": "https://github.com/cran/ADLP", "_commit": { "id": "b09fd60db49dfadb92810318d6710e785432adef", "author": "Yanfeng Li ", "committer": "cran-robot ", "message": "version 0.1.0\n", "time": 1713494068 }, "_maintainer": { "name": "Yanfeng Li", "email": "yanfeng.li@student.unsw.edu.au" }, "_registered": true, "_dependencies": [ { "package": "R", "version": ">= 2.10", "role": "Depends" }, { "package": "methods", "role": "Imports" }, { "package": "knitr", "role": "Suggests" }, { "package": "tidyverse", "role": "Suggests" }, { "package": "SynthETIC", "version": ">= 1.0.0", "role": "Suggests" } ], "_owner": "cran", "_selfowned": false, "_usedby": 0, "_updates": [], "_tags": [], "_stars": 0, "_userbio": { "uuid": 6899542, "type": "organization", "name": "cran", "description": "Unofficial read-only mirror of all CRAN R packages" }, "_downloads": { "count": 211, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/ADLP" }, "_devurl": "https://github.com/agi-lab/adlp", "_searchresults": 0, "_rbuild": "4.6.0", "_assets": [ "extra/ADLP.html", "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/contents.json", "extra/NEWS.html", "extra/NEWS.txt", "extra/readme.html", "extra/readme.md", "manual.pdf" ], "_realowner": "cran", "_cranurl": false, "_releases": [ { "version": "0.1.0", "date": "2024-04-18" } ], "_exports": [ "adlp", "adlp_component", "adlp_components", "adlp_CRPS", "adlp_dens", "adlp_logS", "adlp_partition_ap", "adlp_partition_none", "adlp_simulate", "calc_adlp_component", "calc_adlp_component_lst", "custom_model", "MM_optim", "train_val_split_by_AP", "train_val_split_method1", "train_val_split_method2" ], "_datasets": [ { "name": "test_adlp_component", "title": "Test ADLP Component", "object": "test_adlp_component", "class": [ "adlp_component" ], "fields": [], "table": false, "tojson": false }, { "name": "test_claims_dataset", "title": "Claims Data in data.frame Format", "object": "test_claims_dataset", "class": [ "data.frame" ], "fields": [ "origin", "dev", "calendar", "claims" ], "rows": 1600, "table": true, "tojson": true } ], "_help": [ { "page": "adlp_partition", "title": "Accident and Development period Adjusted Linear Pools partition function", "topics": [ "adlp_partition", "adlp_partition_ap", "adlp_partition_none" ] }, { "page": "calc_adlp_component", "title": "Accident and Development period Adjusted Linear Pools Component Models", "topics": [ "calc_adlp_component", "calc_adlp_component_lst" ] }, { "page": "custom_model", "title": "Custom Model Wrapper", "topics": [ "custom_model", "update.custom_model" ] }, { "page": "MM_optim", "title": "Minorization-Maximisation Algorithm performed to fit the ADLPs", "topics": [ "MM_optim" ] }, { "page": "adlp_func", "title": "Accident and Development period Adjusted Linear Pools (ADLP) Functions", "topics": [ "adlp_CRPS", "adlp_dens", "adlp_func", "adlp_logS", "adlp_simulate", "predict.adlp" ] }, { "page": "adlp", "title": "Accident and Development period Adjusted Linear Pools (ADLP) Models", "topics": [ "adlp", "print.adlp" ] }, { "page": "adlp_component", "title": "Accident and Development period Adjusted Linear Pools Component Models", "topics": [ "adlp_component", "print.adlp_component" ] }, { "page": "adlp_components", "title": "Accident and Development period Adjusted Linear Pools Component Models", "topics": [ "adlp_components", "print.adlp_components" ] }, { "page": "test_adlp_component", "title": "Test ADLP Component", "topics": [ "test_adlp_component" ] }, { "page": "test_claims_dataset", "title": "Claims Data in data.frame Format", "topics": [ "test_claims_dataset" ] }, { "page": "train_val_split", "title": "Train-Validation Split of Claims Triangle", "topics": [ "train_val_split" ] }, { "page": "train_val_split_by_AP", "title": "Train-Validation Split by Accident Period", "topics": [ "train_val_split_by_AP" ] }, { "page": "train_val_split_method1", "title": "Train-Validation Split by Accident Period Method 1", "topics": [ "train_val_split_method1" ] }, { "page": "train_val_split_method2", "title": "Train-Validation Split by Accident Period Method 2", "topics": [ "train_val_split_method2" ] } ], "_readme": "https://github.com/cran/ADLP/raw/HEAD/README.md", "_rundeps": [], "_vignettes": [ { "source": "ADLP-demo.Rmd", "filename": "ADLP-demo.html", "title": "ADLP: Accident and Development period adjusted Linear Pools", "engine": "knitr::rmarkdown", "headings": [ "Set Up", "Training and Validation Split", "Constructing Components", "Fitting ADLPs", "Supported Outputs", "Different partition", "Comparing models through MSE" ], "created": "2024-04-19 02:34:28", "modified": "2024-04-19 02:34:28", "commits": 1 } ], "_score": 2.6989700043360187, "_indexed": true, "_nocasepkg": "adlp", "_universes": [ "cran", "agi-lab" ], "_binaries": [ { "r": "4.7.0", "os": "linux", "version": "0.1.0", "date": "2026-05-25T09:13:41.000Z", "distro": "noble", "commit": "b09fd60db49dfadb92810318d6710e785432adef", "fileid": "eae8ead8101f8be87f526ff49759c08afc9b3e782a32a7139d7cf0f114116fd3", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/26392693516" }, { "r": "4.6.0", "os": "linux", "version": "0.1.0", "date": "2026-05-25T09:13:49.000Z", "distro": "noble", "commit": "b09fd60db49dfadb92810318d6710e785432adef", "fileid": "e30aa42c120446a47bed490c1492ce7290237b7bdf191ad4c435f8205fbddc4f", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/26392693516" }, { "r": "4.6.0", "os": "wasm", "version": "0.1.0", "date": "2026-05-25T09:13:34.000Z", "commit": "b09fd60db49dfadb92810318d6710e785432adef", "fileid": "dd081830440a2bc567744631c5015c0b74e085a032f9330478fda13aa5eb679d", "status": "success", "buildurl": "https://github.com/r-universe/cran/actions/runs/26392693516" } ] }