Figure 9.

Competing models encompassing both possible directions for the causal link between exposure and outcome. (a) $({\mathcal{M}}_{X\to Y})$ Model where the causal relationship is in the expected direction. ${\beta }_{X\to Y}$ is the parameter measuring the (linear) causal effect of the exposure X on the outcome Y. The shaded plate indicates replication across the variants G_j. (b) $({\mathcal{M}}_{Y\to X})$ Alternative model, where the causal relationship is in the reverse direction, from outcome to exposure. Note that ${\beta }_{Y\to X}$ is a different parameter than ${\beta }_{X\to Y}$. The latter is equal to zero for this model, as there is no causal link from X to Y.