Potential Outcomes Flashcards
1
Q
average treatment effect
A
expected value of individual treatment effects
2
Q
why can ATT be different from ATE?
A
treated individuals might not be a random sample of the population
3
Q
how to get the ATT?
A
if selection bias is 0, then ATE = ATT and we get ATT from the difference in means
formally, if the treatment status is independent of potential outcomes, then the conditional expected values are the same as the unconditional
