8 ms·

Observational studies claiming causality are flawed compared to randomised controlled experiments. Hopefully in time research will gravitate toward RCTs. Dist

• mlthoughts2018 8y ago
  But the problem is in many application domains, you cannot obtain holdout sets. For example, if you measure the causal impact of ads for a client, they usually will not agree to accept the cost of holdout inventory in order to measure the effects. They’d rather ensure all the inventory is used, and let the observational post-hoc attempt at causal inference have errors.. they knowingly prefer that trade-off. There are actually a lot of areas like this, where you are just given a one-sided observational data set snd tasked with recovering causal effects with no option to collect a holdout set. Often nobody consciously chose it that way over an RCT, it’s just how it happened.
  • SubiculumCode 8y ago
    or simple covariates like a subject's age. If we could randomly assign people to an age, then...we'd have conquered aging I guess.