Binary Choice under Asymmetric Loss in a Data-Rich Environment: Theory and an Application to Algorithmic Fairness
This addresses the gap in econometrics and machine learning for handling asymmetric loss in binary choice, with an application to fairness in criminal justice, though it appears incremental as it builds on existing methods.
The paper tackles the binary choice problem with asymmetric loss functions in data-rich environments by proposing a simple loss-based reweighting method for logistic regression or machine learning techniques, and applies it to algorithmic fairness in pretrial detentions, showing theoretically valid decisions.
We study the binary choice problem in a data-rich environment with asymmetric loss functions. The econometrics literature covers nonparametric binary choice problems but does not offer computationally attractive solutions in data-rich environments. The machine learning literature has many algorithms but is focused mostly on loss functions that are independent of covariates. We show that theoretically valid decisions on binary outcomes with general loss functions can be achieved via a very simple loss-based reweighting of logistic regression or state-of-the-art machine learning techniques. We apply our analysis to algorithmic fairness in pretrial detentions.