With: Xi Chen , Eric Ghysels, and Rohit Kumar
Abstract:
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 datarich 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 the logistic regression or state-of-the-art machine learning techniques. We apply our analysis to racial justice in pretrial detention. Keywords: binary outcomes, asymmetric losses, machine learning, cost-sensitive classification, pretrial detention.