EconBase
← All papers

What's the Harm? Sharp Bounds on the Fraction Negatively Affected by Treatment

Nathan Kallus

arXiv 20 May 2022 · Statistics — Methodology · 4 citations (OpenAlex)

arXiv:2205.10327 · PDF · DOI · OpenAlex · Extracted main text

Abstract

The fundamental problem of causal inference -- that we never observe counterfactuals -- prevents us from identifying how many might be negatively affected by a proposed intervention. If, in an A/B test, half of users click (or buy, or watch, or renew, etc.), whether exposed to the standard experience A or a new one B, hypothetically it could be because the change affects no one, because the change positively affects half the user population to go from no-click to click while negatively affecting the other half, or something in between. While unknowable, this impact is clearly of material importance to the decision to implement a change or not, whether due to fairness, long-term, systemic, or operational considerations. We therefore derive the tightest-possible (i.e., sharp) bounds on the fraction negatively affected (and other related estimands) given data with only factual observations, whether experimental or observational. Naturally, the more we can stratify individuals by observable covariates, the tighter the sharp bounds. Since these bounds involve unknown functions that must be learned from data, we develop a robust inference algorithm that is efficient almost regardless of how and how fast these functions are learned, remains consistent when some are mislearned, and still gives valid conservative bounds when most are mislearned. Our methodology altogether therefore strongly supports credible conclusions: it avoids spuriously point-identifying this unknowable impact, focusing on the best bounds instead, and it permits exceedingly robust inference on these. We demonstrate our method in simulation studies and in a case study of career counseling for the unemployed.

Citation extraction

63
references
82
in-text mentions
63
distinct cited
7
self-citations
8,723
main-text words

appendix boundary found by appendix_command · 60% 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
1Nathan Kallus (2022) Treatment effect risk: Bounds and inference self0.87462100%
2Maurice Fréchet (1935) Généralisation du théoreme des probabilités totales0.64422100%
3Ludger Rüschendorf (1981) Sharpness of fréchet-bounds0.64422100%
4Luc Behaghel, Bruno Crépon, and Marc Gurgand (2014) Private and public provision of counseling to job seekers: Evidence from a large controlled experiment0.58531100%
5Matteo Bonvini and Edward H Kennedy (2021) Sensitivity analysis via the proportion of unmeasured confounding0.51121100%
6Victor Chernozhukov, Carlos Cinelli, Whitney Newey, Amit Sharma, and… (2021) Omitted variable bias in machine learned causal models0.51121100%
7Jacob Dorn, Kevin Guo, and Nathan Kallus (2021) Doubly-valid/doubly-sharp sensitivity analysis for causal inference with unmeasured confounding self0.51121100%
8Edward H Kennedy (2020) Optimal doubly robust estimation of heterogeneous causal effects0.51121100%
9Nathan Kallus and Angela Zhou (2019) Assessing disparate impacts of personalized interventions: Identifiability and bounds self0.51121100%
10Nathan Kallus, Xiaojie Mao, and Angela Zhou (2021) Assessing algorithmic fairness with unobserved protected class using data combination self0.51121100%

Showing the top 10 of 63 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
1Debiased Machine Learning of Aggregated Intersection Bounds and Other Causal Parameters0.81142
2Limits of Approximating the Median Treatment Effect0.51121
3Doubly-Valid/Doubly-Sharp Sensitivity Analysis for Causal Inference with Unmeasured Confounding0.40511
4Inference on Strongly Identified Functionals of Weakly Identified Functions0.40511
5Policy Learning with Distributional Welfare0.40511
6Predicting the Distribution of Treatment Effects via Covariate-Adjustment, with an Application to Microcredit0.40511
7On the Lower Confidence Band for the Optimal Welfare in Policy Learning0.40511
8Individual Treatment Effect: Prediction Intervals and Sharp Bounds0.40511
9Partial identification via conditional linear programs: estimation and policy learning0.40511
10Causal EpiNets: Precision-corrected Bounds on Individual Treatment Effects using Epistemic Neural Networks0.00011