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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In many investigations, the primary outcome of interest is difficult orexpensive to collect. Examples include long-term health effects of medicalinterventions, measurements requiring expensive testing or follow-up, andoutcomes only measurable on small panels as in marketing. This reduceseffective sample sizes for estimating the average treatment effect (ATE).However, there is often an abundance of observations on surrogate outcomes notof primary interest, such as short-term health effects or online-adclick-through. We study the role of such surrogate observations in theefficient estimation of treatment effects. To quantify their value, we derivethe semiparametric efficiency bounds on ATE estimation with and without thepresence of surrogates and several intermediary settings. The differencebetween these characterizes the efficiency gains from optimally leveragingsurrogates. We study two regimes: when the number of surrogate observations iscomparable to primary-outcome observations and when the former dominates thelatter. We take an agnostic missing-data approach circumventing strongsurrogate conditions previously assumed. To leverage surrogates' efficiencygains, we develop efficient ATE estimation and inference based on flexiblemachine-learning estimates of nuisance functions appearing in the influencefunctions we derive. We empirically demonstrate the gains by studying thelong-term earnings effect of job training.

On the role of surrogates in the efficient estimation of treatment effects with limited outcome data