Paper presents 3 statistical models. Difference is in the confounders: only 1 model contains all pre-defined confounding variables (in DAG). Estimated effect of interest (exposure-outcome) doesn't differ much between models. What will be your take?
#epitwitter
#causalinference
-
-
Replying to @MaartenvSmeden
Need another option - cautious optimism contingent on the model and the outcome
1 reply 0 retweets 3 likes
Replying to @GidMK @MaartenvSmeden
My favorite example of a good outcome in a situation like this was a big study on egg consumption in China. Very convincing model, extremely plausible outcome, I thought the attribution of causality was almost certainly accurate
10:37 PM - 19 Nov 2018
0 replies
0 retweets
1 like
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.