HIGHLIGHTS
SUMMARY
The authors can a priori discard DAG structures where the output variable have a causal influence on the treatment and/or time-of-theday variables (since in the application it is reasonable to expect that the time-of-the-day that the activity was performed, or whether the participant was on- or off-medication can influence the participant`s performance on an activity, but not the other way around). But, most importantly, as fully described in the next section, the conditional independence relations spanned by these 9 models allows the authors to disentangle treatment . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.