This repo contains the code for replicating the figures of the following paper:

Chen, Jiahao, et al. "Fairness Under Unawareness: Assessing Disparity When Protected Class Is Unobserved." Proceedings of the Conference on Fairness, Accountability, and Transparency. ACM, 2019.

The data downloading instruction and data preprocessing instruction are included in the data_cleaning.Rmd. After downloading the data and obtaining the cleaned data "hmda_census_combined.csv", the figures in the paper can be replicated via the code in data_analysis_fat.Rmd. We used a stratified subsample (except the last figure) to speed up the plotting, so the figures may be slightly different from the figures in the paper that use the whole sample.