Dear AllStat
I am doing an analysis of an observational study using the propensity
score approach described by Rosenbaum and Rubin. I have defined a
propensity score with the desired balancing properties and wish
to calculate adjusted event rates using subclassification on the
propensity score. In the paper they calculate the adjusted event rates
and quote a standard error which they say is based on a formula
given in Moteller and Tukey Chapter 11c, Data Analysis and Regression.
My request is this formula.
Many thanks
Andrew
Rosenbaum and Rubin. Reducing bias in observational studies using
subclassification on the propensity score. JASA 1984; 516-524.
----------------------------------
Andrew Charlesworth C.Stat
Head of Statistics
Nottingham Clinical Research Group
[log in to unmask]
Tel: 44 115 9567711
Fax: 44 115 9220960
----------------------------------
|