Science Score: 23.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
✓Committers with academic emails
1 of 12 committers (8.3%) from academic institutions -
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (10.5%) to scientific vocabulary
Repository
scoring rules to evaluate probabilistic forecasts
Basic Info
- Host: GitHub
- Owner: FK83
- Language: R
- Default Branch: master
- Size: 20.4 MB
Statistics
- Stars: 63
- Watchers: 5
- Forks: 16
- Open Issues: 1
- Releases: 1
Metadata Files
readme.md
scoringRules
An R package to compute scoring rules for fixed (parametric) and simulated forecast distributions. Authored by Alexander Jordan (Heidelberg Institute for Theoretical Studies (HITS)), Fabian Krüger (Karlsruhe Institute of Technology (KIT)), Sebastian Lerch (KIT and HITS) and Sam Allen (ETH Zürich), with contributions from Maximiliane Graeter (KIT).
Highlights
- Coherent, dictionary-like reference for computing scoring rules in a wide range of situations
- Previously unavailable closed-form expressions of the CRPS for many parametric distributions
- Efficient implementation thanks to R and Rcpp
- Whenever more than one implementation variant exists, we offer statistically principled default choices
Installation
CRAN version:
r
install.packages("scoringRules")
Development version (GitHub): ```r
install.packages("devtools")
library(devtools) install_github("FK83/scoringRules") ```
Background
Scoring rules are functions S(F, y) which evaluate the accuracy of a forecast distribution F, given that an outcome y was observed. The scoringRules package contains functions to compute scoring rules, for a variety of distributions F that come up in applied work, and several choices of S. Two main classes of distributions are
- Parametric distributions like normal, t, and gamma. For example, most weather forecasts (which apply statistical postprocessing to physical models) take such a form.
- Distributions that are not known analytically, but are indirectly described through a sample of simulaton draws. For example, Bayesian forecasts produced via Markov Chain Monte Carlo (MCMC) take this form.
We cover various scoring rules, including
- the continuous ranked probability score and the logarithmic score
- the energy and variogram scores for multivariate forecast distributions given by discrete samples
- weighted scoring rules for (univariate or multivariate) forecast distributions given by discrete samples
Please refer to the package vignettes 'Evaluating Probabilistic Forecasts with scoringRules' and 'Weighted scoringRules' for details and references.
History
- May 2023: Version 1.1, including threshold and outcome weighted scoring rules (Sam Allen)
- August 2019: Version 1.0.0 on CRAN, vignette published in the Journal of Statistical Software
- November 2017: Version 0.9.4 on CRAN, including a detailed vignette
- July 2017: Vignette for closed-form expressions of the CRPS (Alexander Jordan), and functions for CRPS-based fitting of truncated/censored distributions
- July 7, 2016: Version 0.9 published on CRAN
- September 15, 2014: First commit
GitHub Events
Total
- Issues event: 1
- Watch event: 6
Last Year
- Issues event: 1
- Watch event: 6
Committers
Last synced: over 2 years ago
Top Committers
| Name | Commits | |
|---|---|---|
| aijordan | a@j****e | 234 |
| Fabian Krüger | F****3@g****m | 173 |
| Fabian | f****3@g****m | 68 |
| fabian.krueger | f****r@k****u | 41 |
| SL | s****h@o****m | 35 |
| sallen12 | s****n@s****h | 19 |
| aijordan | a****n | 15 |
| slerch | s****h@g****m | 6 |
| Alexander Jordan | a****n@s****h | 5 |
| onnokleen | m****l@o****e | 1 |
| Ciaran | 4****g | 1 |
| Yihui Xie | x****e@y****e | 1 |
Committer Domains (Top 20 + Academic)
Issues and Pull Requests
Last synced: 11 months ago
All Time
- Total issues: 10
- Total pull requests: 52
- Average time to close issues: 8 months
- Average time to close pull requests: about 14 hours
- Total issue authors: 9
- Total pull request authors: 6
- Average comments per issue: 1.9
- Average comments per pull request: 0.06
- Merged pull requests: 50
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 1
- Pull requests: 2
- Average time to close issues: N/A
- Average time to close pull requests: 26 minutes
- Issue authors: 1
- Pull request authors: 1
- Average comments per issue: 0.0
- Average comments per pull request: 0.0
- Merged pull requests: 2
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
- bastistician (2)
- dmi3kno (1)
- kashif (1)
- Ostovane (1)
- ghurault (1)
- hrlai (1)
- groceryheist (1)
- ciaran-g (1)
- mbojan (1)
Pull Request Authors
- aijordan (24)
- slerch (19)
- sallen12 (8)
- onnokleen (1)
- yihui (1)
- ciaran-g (1)
Top Labels
Issue Labels
Pull Request Labels
Packages
- Total packages: 1
-
Total downloads:
- cran 3,257 last-month
- Total docker downloads: 43,473
- Total dependent packages: 8
- Total dependent repositories: 14
- Total versions: 13
- Total maintainers: 1
cran.r-project.org: scoringRules
Scoring Rules for Parametric and Simulated Distribution Forecasts
- Homepage: https://github.com/FK83/scoringRules
- Documentation: http://cran.r-project.org/web/packages/scoringRules/scoringRules.pdf
- License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
-
Latest release: 1.1.3
published almost 2 years ago
Rankings
Maintainers (1)
Dependencies
- R >= 3.00 depends
- MASS * imports
- Rcpp >= 0.12.0 imports
- knitr * imports
- methods * imports
- crch * suggests
- gsl >= 1.8 suggests
- hypergeo >= 1.0 suggests
- rmarkdown * suggests
- testthat * suggests