City Research Online

Abstract

Different notions of fairness and discrimination have been extensively discussed in the machine learning and insurance pricing literatures. As not all fairness criteria can be concurrently satisfied, it is important to develop metrics that allow the assessment of materiality of discriminatory effects and the trade-offs between various criteria. Methods from sensitivity analysis have been deployed for the measurement of demographic unfairness, that is, the statistical dependence of risk predictions on protected attributes. We produce a sensitivity-based measure for the different phenomenon of proxy discrimination, referring to the implicit inference of protected attributes from other covariates. For this, we first define a set of admissible prices that avoid proxy discrimination. Then, the measure is defined as the normalised L2 -distance of a price from the closest element in that set. Furthermore, we consider the attribution of theproxy discrimination measure to individual (or subsets of) covariates and investigate how properties of the data generating process are reflected in those metrics. Finally, we build on the global (i.e., portfolio-wide) measures of demographic unfairness and proxy discrimination to propose local (i.e., policyholder-specific) measures, which allow a fine-grained understanding of discriminatory effects across a collective of policyholders.

Publication Type:	Monograph (Working Paper)
Additional Information:	Copyright the authors, 2024.
Publisher Keywords:	Proxy discrimination, demographic parity, global sensitivity analysis, insurance pricing, algorithmic fairness
Subjects:	H Social Sciences > HF Commerce H Social Sciences > HN Social history and conditions. Social problems. Social reform
Departments:	Bayes Business School Bayes Business School > Actuarial Science & Insurance
SWORD Depositor:	Symplectic Administrator

Export

Downloads