average treatment effect

Uses your system voice; voice and rate in settings.

Q4828276Snapshot

≈ 5 min read

Names in other languages

  • DeutschDurchschnittlicher Behandlungseffekt
  • PortuguêsEfeito médio do tratamento
  • Englishaverage treatment effect
  • eoaveraĝa efiko de trakto
  • УкраїнськаСередній ефект впливу
  • 한국어평균 처치 효과
  • 日本語平均処置効果
  • 中文平均处理效应

Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.

The average treatment effect (ATE) is a measure used to compare treatments (or interventions) in randomized experiments, evaluation of policy interventions, and medical trials. The ATE measures the difference in mean (average) outcomes between units assigned to the treatment and units assigned to the control. In a randomized trial (i.e., an experimental study), the average treatment effect can be estimated from a sample using a comparison in mean outcomes for treated and untreated units.

How it works

However, the ATE is generally understood as a causal parameter (i.e., an estimate or property of a population) that a researcher desires to know, defined without reference to the study design or estimation procedure. Both observational studies and experimental study designs with random assignment may enable one to estimate an ATE in a variety of ways. The average treatment effect is under some conditions directly related to the partial dependence plot.

Evidence and sources

The average treatment effect (ATE) is a measure used to compare treatments (or interventions) in randomized experiments, evaluation of policy interventions, and medical trials. The ATE measures the difference in mean (average) outcomes between units assigned to the treatment and units assigned to the control. In a randomized trial (i.e., an experimental study), the average treatment effect can be estimated from a sample using a comparison in mean outcomes for treated and untreated units.

However, the ATE is generally understood as a causal parameter (i.e., an estimate or property of a population) that a researcher desires to know, defined without reference to the study design or estimation procedure. Both observational studies and experimental study designs with random assignment may enable one to estimate an ATE in a variety of ways. The average treatment effect is under some conditions directly related to the partial dependence plot. The treatment effect for individual i is given by y_{1}(i)-y_{0}(i)=\beta(i) .

Impact

In the general case, there is no reason to expect this effect to be constant across individuals. The average treatment effect is given by \text{ATE} = \mathbb{E}[y_{1}-y_{0}] and can be estimated (if a law of large numbers holds) \widehat{ATE} = \frac{1}{N}\sum_i (y_{1}(i)-y_{0}(i)) where the summation occurs over all N individuals in the population.

If we could observe, for each individual, y_{1}(i) and y_{0}(i) among a large representative sample of the population, we could estimate the ATE simply by taking the average value of y_{1}(i)-y_{0}(i) across the sample. However, we can not observe both y_{1}(i) and y_{0}(i) for each individual since an individual cannot be both treated and not treated. For example, in the drug example, we can only observe y_{1}(i) for individuals who have received the drug and y_{0}(i) for those who did not receive it.

This is the main problem faced by scientists in the evaluation of treatment effects and has triggered a large body of estimation techniques. Some researchers call a treatment effect "heterogenous" if it affects different individuals differently (heterogeneously). For example, perhaps the above treatment of a job search monitoring policy affected men and women differently, or people who live in different states differently.

ATE requires a strong assumption known as the stable unit treatment value assumption (SUTVA) which requires the value of the potential outcome y(i) be unaffected by the mechanism used to assign the treatment and the treatment exposure of all other individuals. Let d be the treatment, the treatment effect for individual i is given by y_{1}(i,d)-y_{0}(i,d) . The SUTVA assumption allows us to declare y_{1}(i,d) = y_{1}(i), y_{0}(i,d)=y_{0}(i) .

One way to look for heterogeneous treatment effects is to divide the study data into subgroups (e.g., men and women, or by state), and see if the average treatment effects are different by subgroup. If the average treatment effects are different, SUTVA is violated. A per-subgroup ATE is called a "conditional average treatment effect" (CATE), i.e. the ATE conditioned on membership in the subgroup. CATE can be used as an estimate if SUTVA does not hold.

A challenge with this approach is that each subgroup may have substantially less data than the study as a whole, so if the study has been powered to detect the main effects without subgroup analysis, there may not be enough data to properly judge the effects on subgroups. There is some work on detecting heterogeneous treatment effects using random forests as well as detecting heterogeneous subpopulations using cluster analysis.

Recently, metalearning approaches have been developed that use arbitrary regression frameworks as base learners to infer the CATE. Representation learning can be used to further improve the performance of these methods.

Text derived from https://en.wikipedia.org/w/index.php?oldid=1357740478 under CC BY-SA 4.0; modifications have been made.

Provenance

snapshot average-treatment-effect-551456-314a8721ef00c846 · hash 314a8721ef00c846c90d3dfc857a3c9f065e7472f8cae4798b476869a84f99db · captured 1780566970.0 · version history

Cite this page as: encyclopegia.org, “average treatment effect”, snapshot average-treatment-effect-551456-314a8721ef00c846. https://encyclopegia.org/a/average-treatment-effect

Sources cited in this revision

No external citation was extracted for this revision. That is a statement about the source, not a claim that none exists.