The R
package csranks
provides statistical tools for estimation and inference involving ranks (the position in a ranking). Two central functions are csranks
for confidence sets for ranks and lmranks
for regressions involving ranks, e.g. rank-rank regressions that are popular in applied work in economics.
The functions are based on recent work developing these procedures and their theoretical properties. The confidence sets for ranks are based on Mogstad, Romano, Shaikh, and Wilhelm (2023) and Bazylik, Mogstad, Romano, Shaikh, and Wilhelm (2022). The inference methods for regressions involving ranks are developed in Chetverikov and Wilhelm (2023).
You can install the released version of csranks
from CRAN with:
install.packages("csranks")
You can install the development version of csranks
from GitHub with:
# install.packages("remotes")
remotes::install_github("danielwilhelm/R-CS-ranks")
If you want to play around with the confidence sets for ranks, feel free to try out demo available here.
Here is a sneak peek: