English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In policy research, one of the most critical analytic tasks is to estimatethe causal effect of a policy-relevant shift to the distribution of acontinuous exposure/treatment on an outcome of interest. We call this problemshift-response function (SRF) estimation. Existing neural network methodsinvolving robust causal-effect estimators lack theoretical guarantees andpractical implementations for SRF estimation. Motivated by a keypolicy-relevant question in public health, we develop a neural network methodand its theoretical underpinnings to estimate SRFs with robustness andefficiency guarantees. We then apply our method to data consisting of 68million individuals and 27 million deaths across the U.S. to estimate thecausal effect from revising the US National Ambient Air Quality Standards(NAAQS) for PM 2.5 from 12 $\mu g/m^3$ to 9 $\mu g/m^3$. This change has beenrecently proposed by the US Environmental Protection Agency (EPA). Our goal isto estimate, for the first time, the reduction in deaths that would result fromthis anticipated revision using causal methods for SRFs. Our proposed method,called {T}argeted {R}egularization for {E}xposure {S}hifts with Neural{Net}works (TRESNET), contributes to the neural network literature for causalinference in two ways: first, it proposes a targeted regularization loss withtheoretical properties that ensure double robustness and achieves asymptoticefficiency specific for SRF estimation; second, it enables loss functions fromthe exponential family of distributions to accommodate non-continuous outcomedistributions (such as hospitalization or mortality counts). We complement ourapplication with benchmark experiments that demonstrate TRESNET's broadapplicability and competitiveness.

Causal Estimation of Exposure Shifts with Neural Networks: Evaluating the Health Benefits of Stricter Air Quality Standards in the US