B1399
Title: Model parameterisations using causal quantities
Authors: Robin Evans - University of Oxford (United Kingdom) [presenting]
Vanessa Didelez - Leibniz Institute for Prevention Research and Epidemiology - BIPS, University of Bremen (Germany)
Abstract: Many causal parameters of interest are marginal quantities: that is, they are formed by averaging over a real or hypothetical population. Several authors have noted the practical difficulties of dealing with such quantities. This is due to the apparent incompatibility of a marginal parameterisation involving the causal quantity of interest and conditional parametric models used for dealing with confounding (either observed or unobserved). In some cases, the so-called $g$-null paradox implies that it is logically impossible for the conditional models and the marginal null hypothesis to hold simultaneously. This means that even simulating from the null model to test new methods is not always possible, let alone performing likelihood-based inference. We adapt a previous marginal parameterisations to causal models, allowing us to parameterise a wide range of causal models including marginal structural models (MSMs), Cox MSMs and structural nested models. This makes it easy to simulate from and fit models; to introduce possibly high-dimensional individual-level covariates; and to include additional assumptions such as stationarity or symmetry. Our approach also avoids the null paradox where possible, and makes transparent when it is unavoidable by assumption.
Journal