For two response variables yt and yc corresponding to two treatments (or two policies) T and C, we wish to learn about quantiles of y1 - yc from the marginal quantiles of yt and yc; only one of yt and yc is observed for an individual. We find that, in general, this is difficult for quantiles other than the median unless strong assumptions are imposed on how yt is related to yc. For the median, we present conditions under which the sign of the median treatment effect is identified.
|Number of pages||10|
|Journal||Journal of the Royal Statistical Society. Series B: Statistical Methodology|
|Publication status||Published - 2000 Dec 1|
- Treatment effect
ASJC Scopus subject areas
- Statistics and Probability