Causal inference with interfering units for cluster and population level intervention programs

Seminar | November 8 | 4-5 p.m. | 1011 Evans Hall

 Fabrizia Mealli, University of Florence

 Department of Statistics

Interference arises when an individual's potential outcome
depends on the individual treatment and also on the treatment
of others. A common assumption in the causal inference literature in the
presence of interference is partial interference, implying that the
population can be partitioned in clusters of units whose potential
outcomes only depend on the treatment of other units within the same
Previous literature has defined average potential outcomes under
counterfactual scenarios where treatment is randomly allocated to units
within a cluster with equal probability. However, within clusters there
may be units that are more or less likely to receive treatment based on
covariates or neighbors' treatment. We define estimands that describe
average potential outcomes for realistic regimes taking into consideration
the units' covariates, as well as dependence between units' treatment
assignment. We discuss these estimands, propose unbiased estimators and
derive asymptotic results as the number of clusters grows. Finally, we
estimate effects in a comparative effectiveness study of emission
reduction technologies on ambient ozone concentration in the presence of
interference. Joint work with Georgia Papadogeorgou and Corwin Zigler.

 (510) 642-2781