shamindras / maars

An R implementation of Models As Approximations
https://shamindras.github.io/maars/
Other
11 stars 1 forks source link
linear-regression ordinary-least-squares r statistics

Codecov test
coverage R build
status MIT
License

maars - an R implementation of Models As Approximations

The goal of the maars package is to implement the Models As Approximations series of statistics papers (Buja, Brown, Berk, et al. 2019) and (Buja, Brown, Kuchibhotla, et al. 2019). This package was inspired by the fantastic series of lectures by Prof. Arun Kumar Kuchibhotla and Prof. Andreas Buja, as part of the “STAT 36761: Modern Linear Regression” course at Carnegie Mellon University (CMU) in Fall 2020.

Installation and User Guide

To get a bug fix or to use a feature from the development version, you can install the development version of maars from GitHub, as follows:

# install.packages("devtools")
devtools::install_github("shamindras/maars")

More detailed instructions and user guides can be found at the official package website. The source code for the maars package can be found on github.

Citation

If you are in R you can simply run the following command to get the BibTeX citation for maars:

citation("maars")

Alternatively, please use the following BibTeX citation:

@misc{fogliato2021maars,
  title  = {maars: Tidy Inference under the 'Models as Approximations' Framework in R},
  author = {Riccardo Fogliato and Shamindra Shrotriya and Arun Kumar Kuchibhotla},
  year   = {2021},
  eprint = {arXiv:2106.11188},
  url    = {https://shamindras.github.io/maars/},
  note   = {R package version 0.3.0}
}

Code of Conduct

Please note that the maars project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

Related Packages

While maars has it’s own approach and API for performing valid inference under model misspecification for OLS, it may not meet your particular needs. Here is a listing of other leading R packages in this field which you may want to try, with links to their project pages (listed alphabetically):

Credits

This package is developed and maintained by:

We want this to be a community project, so please feel free to contact us, or file an issue if you would like to contribute to it.

References

Buja, Andreas, Lawrence Brown, Richard Berk, Edward George, Emil Pitkin, Mikhail Traskin, Kai Zhang, and Linda Zhao. 2019. “Models as Approximations I: Consequences Illustrated with Linear Regression.” *Statist. Sci.* 34 (4): 523–44.
Buja, Andreas, Lawrence Brown, Arun Kumar Kuchibhotla, Richard Berk, Edward George, and Linda Zhao. 2019. “Models as Approximations II: A Model-Free Theory of Parametric Regression.” *Statist. Sci.* 34 (4): 545–65.