Unifying Python framework for evaluating robustness of pairwise comparison methods against rank reversal
Just run robustness.py (depends on config.py) in order to replicate the results from "Jan Górecki, David Bartl and Jaroslav Ramík (2023). Robustness of priority deriving methods for pairwise comparison matrices against rank reversal: A probabilistic approach. Annals of Operations Research. DOI: https://doi.org/10.1007/s10479-023-05753-0"
If one does not want to compute all the results but wants to replicate the plots, just switch the value of argument load_results to True in replicate(), and the precomputed results from the folder results will be used.