fastfrechet

fastfrechet: An R package for fast implementation of Fréchet regression with distributional responses - Published in JOSS (2025)

https://github.com/alexandercoulter/fastfrechet

Repository

README.Rmd

---
output: github_document
---



```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

# fastfrechet: Fast Fréchet regression with univariate distributional responses


[![R-CMD-check](https://github.com/alexandercoulter/fastfrechet/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/alexandercoulter/fastfrechet/actions/workflows/R-CMD-check.yaml)
[![Codecov test coverage](https://codecov.io/gh/alexandercoulter/fastfrechet/graph/badge.svg)](https://app.codecov.io/gh/alexandercoulter/fastfrechet)


`fastfrechet` is an `R` package for performing Fréchet regression with univariate distributional responses, including variable selection. The corresponding methodological frameworks have been described in

* [Petersen A and Müller H-G (2019), "Fréchet regression for random objects with Euclidean predictors", *Annals of Statistics*](https://doi.org/10.1214/17-AOS1624)
* [Tucker DC, Wu Y and Müller H-G (2023), "Variable Selection for Global Fréchet Regression", *JASA*](https://doi.org/10.1080/01621459.2021.1969240)

The fast implementation and stability selection is based on the algorithm developed in 

* [Coulter A, Aurora RN, Punjabi N and Gaynanova I (2024), “Fast variable selection for distributional regression with application to continuous glucose monitoring data.” *arXiv*](https://arxiv.org/abs/2403.00922)

## Installation

To use `fastfrechet`, you need to install [`R`](https://cran.r-project.org/). To enhance your user experience, you may use some IDE for it (e.g. [`RStudio`](https://www.rstudio.com/)).

The development version of [`fastfrechet`](https://github.com/alexandercoulter/fastfrechet) is available on GitHub. You can download it with the help of the `devtools` package in `R` as follow:

```{r, eval = FALSE}
install.packages("devtools")
devtools::install_github("https://github.com/alexandercoulter/fastfrechet", build_vignettes = TRUE)
```

## Community Guidelines


1.  Contributions and suggestions to the software are always welcome.
    Please consult our [contribution guidelines](.github/CONTRIBUTING.md) prior
    to submitting a pull request.
2.  Report issues or problems with the software using github’s [issue tracker](https://github.com/alexandercoulter/fastfrechet/issues).
3.  Contributors must adhere to the [Code of Conduct](CODE_OF_CONDUCT.md).

Owner

JOSS Publication

fastfrechet: An R package for fast implementation of Fréchet regression with distributional responses

Published

May 01, 2025

DOI

10.21105/joss.07925

Volume 10, Issue 109, Page 7925

Authors

Alexander Coulter
Department of Statistics, Texas A&M University, United States

Rebecca Lee
Department of Statistics, Texas A&M University, United States

Irina Gaynanova
Department of Biostatistics, University of Michigan, United States

Editor

Fabian Scheipl

Tags

Fréchet regression Variable selection Distributions Wasserstein Metric space

View PDF Review Thread Software Archive

GitHub Events

Issues and Pull Requests

Last synced: 10 months ago