I have read that matching nearly always yields the $ATT$ effect, but that subclassification matching can yield the $ATE$. I am therefore wondering what is a heuristic for determining what kinds of matching lead to $ATE$ vs $ATT$ effects? For example, what does exact matching, propensity score matching, or genetic matching give?
$\begingroup$
$\endgroup$
2
-
4$\begingroup$ Did you read Greifer & Stuart (2023), which directly answers this question, which I cited in my answer to your very similar previous question? $\endgroup$– NoahCommented May 14 at 12:53
-
3$\begingroup$ Noah's paper (linked above) answers this question in great detail. If there are parts of the paper you're struggling to understand, I'd suggest you ask a specific question and reference the section you're not getting. $\endgroup$– num_39Commented Sep 2 at 10:53
Add a comment
|