EconBase
← All papers

Assessing Disparate Impacts of Personalized Interventions: Identifiability and Bounds

Nathan Kallus, Angela Zhou

arXiv 4 Jun 2019 · Statistics — Machine Learning · 8 citations (OpenAlex)

arXiv:1906.01552 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Personalized interventions in social services, education, and healthcare leverage individual-level causal effect predictions in order to give the best treatment to each individual or to prioritize program interventions for the individuals most likely to benefit. While the sensitivity of these domains compels us to evaluate the fairness of such policies, we show that actually auditing their disparate impacts per standard observational metrics, such as true positive rates, is impossible since ground truths are unknown. Whether our data is experimental or observational, an individual's actual outcome under an intervention different than that received can never be known, only predicted based on features. We prove how we can nonetheless point-identify these quantities under the additional assumption of monotone treatment response, which may be reasonable in many applications. We further provide a sensitivity analysis for this assumption by means of sharp partial-identification bounds under violations of monotonicity of varying strengths. We show how to use our results to audit personalized interventions using partially-identified ROC and xROC curves and demonstrate this in a case study of a French job training dataset.

Citation extraction

53
references
74
in-text mentions
54
distinct cited
0
self-citations
6,179
main-text words

appendix boundary found by appendix_command · 67% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1N. Kallus and A. Zhou (2019) The fairness of risk scores beyond classification: Bipartite ranking and the xauc metric1.00053100%
2M. Hardt, E. Price, N. Srebro, et al (2016) Equality of opportunity in supervised learning0.84333100%
3S. Wager and S. Athey (2017) Estimation and inference of heterogeneous treatment effects using random forests0.84333100%
4T. Kitagawa and A. Tetenov (2015) Empirical welfare maximization0.7373367%
5C. F. Manski (2003) Partial identification of probability distributions0.73732100%
6S. Barocas, M. Hardt, and A. Narayanan (2018) Fairness and Machine Learning0.64422100%
7J. L. Hill (2011) Bayesian nonparametric modeling for causal inference0.64422100%
8S. Wager and S. Athey (2017) Efficient policy learning0.64422100%
9M. Adler Well-Being and Fair Distribution0.5112250%
10M. Berger, D. A. Black, and J. A. Smith (2000) Econometric evaluation of labour market policies, chapter EvaluatingProfiling as a Means of Allocating Government Service, pages…0.5112250%

Showing the top 10 of 54 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1What's the Harm? Sharp Bounds on the Fraction Negatively Affected by Treatment0.51121
2Debiased Machine Learning of Set-Identified Linear Models0.40511
3Generalized Lee Bounds0.40511
4Treatment Effect Risk: Bounds and Inference0.40511
5Debiased Machine Learning of Aggregated Intersection Bounds and Other Causal Parameters0.40511