ABSTRACT
In this section I want to consider a situation that would usually be described by saying that x3 causes x2 which then goes on to cause x\. In the preceding section I have dealt with some of the complexities of what this means, and so in this section I want to take a simpler view. I assume that the causal diagram of Figure 1 shows all of the relevant causal relationships, and that the IFR holds, so that none of the parameters shown in this figure are miss-specified. This is rather a lot of assumption, but perhaps one of the lessons of preceding chapters is that if we shrink from making any assumptions then we are likely to succumb to analysis paralysis.
