Search
Program Calendar
Browse By Day
Browse By Unit
Browse By Session Submission Type
Browse By Descriptors
Help
About Vancouver
Personal Schedule
Sign In
To identify comparable individuals in nonexperimental studies recent literature has brought forth the prognosis score (PG) as a complement to the propensity score (PS). Similar to PSs, PGs collapse covariates into a unidimensional measure summarizing covariates' associations with the potential outcomes. The pairing of PSs and PGs has largely come through a unified distance measure (Mahalanobis). In this paper, we discuss why measures which unify PGs and PSs may be suboptimal and present an approach which uses these scores independently. Results suggest that PS matching within strata defined by PGs outperforms both matching on the PS only as well as matching on PSs and PGs via a unified distance such as Mahalanobis.