Skip to main content
. Author manuscript; available in PMC: 2012 Sep 1.
Published in final edited form as: Stat Biosci. 2011 Sep;3(1):6–27. doi: 10.1007/s12561-011-9033-6

Fig. 1.

Fig. 1

(a) Estimated propensity for choosing BCSRT among BCSRT and MST receivers. (b) Heterogeneity in treatment effects across deciles (ηq, q = 1, …, 10) of 2. (c) Heterogeneity in treatment effects across UD, propensity to select BCSRT based on unobserved confounders. (Solid circles in (b) and (c) represent treatment effects that are significant at 5% level)