Closed malcolmbarrett closed 2 years ago
Looks like this is relevant to #41 @LucyMcGowan
We should come up with corresponding questions in the disney data. Some targets won't make much sense but that's probably helpful to see. We should also include the related functions from matchit and propensity
Ok, we have examples in the first part of the chapter. I also think we should have examples of what the question might be for the touringplans data (from Disney's perspective, as the ones who can set the intervention). Travis also suggests we talk about how, from the perspective of a rider, prediction modeling might be useful, connecting back to chapter 1
ATO: There are probably days that will NEVER be EMH and somedays that always would
Based on https://arxiv.org/ftp/arxiv/papers/2106/2106.10577.pdf