rankrate: Joint Statistical Models for Preference Learning with Rankings and Ratings

This package allows for joint modeling of ranking and rating preference data via the Mallows-Binomial model (Pearce and Erosheva 2022a). Functions in the package may be used for density calculation, random data generation, and fitting the Mallows-Binomial model to data via multiple exact and approximate methods. Uncertainty quantification and estimation of confidence intervals is also possible via the nonparametric bootstrap, whose asymptotic validity was proven in Pearce and Erosheva (2022b). Additionally, the package includes 3 “toy” data sets and 1 real data from the American Institute of Biological Sciences, which were all studied in Gallo et al. (2023).

For more details on how to use this package, see the tutorial.

A published version of the package may be installed from CRAN, or a development version from Github for the most up-to-date functionality:

## Published (CRAN) version
install.packages("rankrate")  

## Development (Github) version
# install.packages("devtools") # uncomment if you haven't installed 'devtools' before
devtools::install_github("pearce790/rankrate")

After installation, load the package with the following code:

library(rankrate)

References

Gallo, Stephen A, Michael Pearce, Carole J Lee, and Elena A Erosheva. 2023. “A New Approach to Peer Review Assessments: Score, Then Rank.” Research Integrity and Peer Review 8 (10): 10.
Pearce, Michael, and Elena A Erosheva. 2022a. “A Unified Statistical Learning Model for Rankings and Scores with Application to Grant Panel Review.” Journal of Machine Learning Research 23 (210): 1–33.
———. 2022b. “On the Validity of Bootstrap Uncertainty Estimates in the Mallows-Binomial Model.” arXiv Preprint arXiv:2206.12365.