An official website of the United States government
We study regressions with multiple treatments and a set of controls that is flexible enough to purge omitted variable bias. We show these regressions generally fail to estimate convex averages of heterogeneous treatment effects—instead, estimates of each treatment’s effect are contaminated by nonconvex averages of the effects of other treatments. We discuss three estimation approaches that avoid such contamination bias, including the targeting of easiest-to-estimate weighted average effects. A reanalysis of nine empirical applications finds economically and statistically meaningful contamination bias in observational studies; contamination bias in experimental studies is more limited due to smaller variability in propensity scores.
more »
« less
Presumptive Contamination: A New Approach to PFAS Contamination Based on Likely Sources
- Award ID(s):
- 1827817
- PAR ID:
- 10373557
- Publisher / Repository:
- American Chemical Society
- Date Published:
- Journal Name:
- Environmental Science & Technology Letters
- Volume:
- 9
- Issue:
- 11
- ISSN:
- 2328-8930
- Page Range / eLocation ID:
- p. 983-990
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Many machine learning problems can be characterized by mutual contamination models. In these problems, one observes several random samples from different convex combinations of a set of unknown base distributions and the goal is to infer these base distributions. This paper considers the general setting where the base distributions are defined on arbitrary probability spaces. We examine three popular machine learning problems that arise in this general setting: multiclass classification with label noise, demixing of mixed membership models, and classification with partial labels. In each case, we give sufficient conditions for identifiability and present algorithms for the infinite and finite sample settings, with associated performance guarantees.more » « less
-
ABSTRACT Contamination offers a new observatory for anthropological theory. But does it bring us closer to the world at hand? I have spent the past five years working with residents in Bennington, Vermont, and Hoosick Falls, New York, in pursuit of justice after the toxin PFOA was discovered in their drinking water. Turning from advocacy to writing, I've been struck by how prominent toxicity is becoming in certain currents of anthropological theory and how little those theories illuminate about the protests against contamination I participated in. As the theoretical dazzle of contamination surges forward toward experimental futures, planetary futures, and queer futures, toxicity can become an oracle whose ethnographic significance lies more in its prophetic intimation than in its present inhabitation. Staying close to the experience of a New England community protesting industrial pollution, I show how the ethnographic realities of contamination can orient theory for a better world without first resigning us to the loss of the present. [toxics,materiality,futures,environmental justice,PFAS,plastic pollution,United States]more » « less
