Designing Experiments Toward Shrinkage Estimation

by   Evan T. R. Rosenman, et al.

We consider how increasingly available observational data can be used to improve the design of randomized controlled trials (RCTs). We seek to design a prospective RCT, with the intent of using an Empirical Bayes estimator to shrink the causal estimates from our trial toward causal estimates obtained from an observational study. We ask: how might we design the experiment to better complement the observational study in this setting? We propose using an estimator that shrinks each component of the RCT causal estimator toward its observational counterpart by a factor proportional to its variance. First, we show that the risk of this estimator can be computed efficiently via numerical integration. We then propose algorithms for determining the best allocation of units to strata (the best "design"). We consider three options: Neyman allocation; a "naive" design assuming no unmeasured confounding in the observational study; and a "defensive" design accounting for the imperfect parameter estimates we would obtain from the observational study with unmeasured confounding. We also incorporate results from sensitivity analysis to establish guardrails on the designs, so that our experiment could be reasonably analyzed with and without shrinkage. We demonstrate the superiority of these experimental designs with a simulation study involving causal inference on a rare, binary outcome.


page 1

page 2

page 3

page 4


Designing Experiments Informed by Observational Studies

The increasing availability of passively observed data has yielded a gro...

Removing Hidden Confounding by Experimental Grounding

Observational data is increasingly used as a means for making individual...

Absolute and Relative Bias in Eight Common Observational Study Designs: Evidence from a Meta-analysis

Observational studies are needed when experiments are not possible. With...

Combining Observational and Experimental Datasets Using Shrinkage Estimators

We consider the problem of combining data from observational and experim...

Conceptualising Natural and Quasi Experiments in Public Health

Background: Natural or quasi experiments are appealing for public health...

When can relative risks provide causal estimates?

It is emphasised that for epidemiological studies where disease incidenc...

Assessment of Immune Correlates of Protection via Controlled Vaccine Efficacy and Controlled Risk

Immune correlates of protection (CoPs) are immunologic biomarkers accept...

Please sign up or login with your details

Forgot password? Click here to reset