BRcal: An R Package to Boldness-Recalibrate Probability Predictions

TL;DR

BRcal is an R package that enables users to responsibly adjust probability predictions to be bolder or more cautious while maintaining calibration, supported by a novel optimization framework and practical visualization tools.

Contribution

The paper introduces a new boldness-recalibration method formulated as a nonlinear optimization problem, implemented in the BRcal R package with visualization and real-world case study.

Findings

Provides a flexible control of calibration-boldness tradeoff.

Maximizes posterior probability of calibration via linear log odds likelihood.

Demonstrates effectiveness through a housing foreclosure case study.

Abstract

When probability predictions are too cautious for decision making, boldness-recalibration enables responsible emboldening while maintaining the probability of calibration required by the user. We formulate boldness-recalibration as a nonlinear optimization of boldness with a nonlinear inequality constraint on calibration. We further show that recalibration based on the maximized linear log odds likelihood also maximizes the posterior probability of calibration. We introduce BRcal, an R package implementing boldness-recalibration and supporting methodology as recently proposed. The BRcal package provides direct control of the calibration-boldness tradeoff and visualizes how different calibration levels change individual predictions. We present a new real world case study involving housing foreclosure predictions. The BRcal package is available on the Comprehensive R Archive Network (CRAN) (https://cran.r-project.org/web/packages/BRcal/index.html) and on Github (https://github.com/apguthrie/BRcal).