David van Dyk, Ph. D.
Department of Statistics
Harvard University
Wednesday, February 19, 2003
3:30 PM
Mayo C231
Minneapolis Campus
Abstract:
In this talk, we develop the theoretical properties of the propensity function
which is a generalization of the propensity score of Rosenbaum and Rubin (Biometrika,
1983). Methods based on the propensity score have long been used for causal
inference in observational studies; they are easy to use and can effectively
reduce the bias caused by non-random treatment assignment. Although treatment
regimes are often not binary in practice, the propensity score methods are generally
confined to binary treatment scenarios. Two possible exceptions were suggested
by Joffe and Rosenbaum (Am. J. of Epidemiology, 1999) and Imbens (Biometrika,
2000) for ordinal and categorical treatments, respectively. In this talk, we
develop theory and methods which encompass all of these techniques and widen
their applicability by allowing for arbitrary treatment regimes. We illustrate
our propensity function methods by applying them to two data sets; we estimate
the effect of smoking on medical expenditure and the effect of schooling on
wages. We also conduct Monte Carlo experiments to investigate the performance
of our methods.
This is joint work with Kosuke Imai, Department of Government, Harvard University.