https://github.com/alexisderumigny/mmdcopula
Robust Estimation of Copulas by Maximum Mean Discrepancy
https://github.com/alexisderumigny/mmdcopula
copula-models copulas mmd r-package r-pkg
Last synced: 8 months ago
JSON representation
Robust Estimation of Copulas by Maximum Mean Discrepancy
- Host: GitHub
- URL: https://github.com/alexisderumigny/mmdcopula
- Owner: AlexisDerumigny
- Created: 2020-10-01T12:53:14.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2022-04-25T10:23:55.000Z (about 4 years ago)
- Last Synced: 2025-10-11T17:51:30.595Z (8 months ago)
- Topics: copula-models, copulas, mmd, r-package, r-pkg
- Language: R
- Homepage:
- Size: 158 KB
- Stars: 5
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Package MMDCopula
===================
This package implements the robust estimation procedure of copulas via maximum mean discrepancy (MMD),
following the article Alquier, Chérief-Abdellatif, Derumigny, and Fermanian, J.D. (2022),
Estimation of copulas via Maximum Mean Discrepancy.
*Journal of the American Statistical Association*, [doi:10.1080/01621459.2021.2024836](https://doi.org/10.1080/01621459.2021.2024836).
**How to install**
The release version on CRAN:
```r
install.packages("MMDCopula")
```
The development version from GitHub:
```r
# install.packages("remotes")
remotes::install_github("AlexisDerumigny/MMDCopula")
```
**Main functions**
* `BiCopEstMMD`: estimate the parameter of a parametric bivariate copula by MMD minimization.
* `BiCopConfIntMMD`: compute a bootstrap-based or subsampling-based confidence interval for the parameter of a parametric bivariate copula.
* `BiCopGradMMD`: compute the gradient of the MMD criteria. Used in `BiCopEstMMD`.
**Functions for simulation and inference for the Marshall-Olkin copula**
* `BiCopSim.MO`: simulation of observations following a Marshall-Olkin copula.
* `BiCopEst.MO`: estimation of the parameter of a Marshall-Olkin copula.
* `BiCopPar2Tau.MO` and `BiCopTau2Par.MO`: convert between the parameter and the Kendall's tau of a Marshall-Olkin copula.
**Other functions**
* `BiCopParamDistLp`: compute the $L^p$ distance between two parametric copula models.