SEMINAR

Bias Reduction versus Sufficient Dimension Reduction


David Nelson
Center for Chronic Disease
Outcome Research
Minneapolis VA Medical Center


Wednesday, May 2, 2007
3:30pm
Moos 2-690
Minneapolis Campus


Abstract:

Propensities and related balancing scores play a central role in bias reduction. We review the foundations of Propensity Theory and discuss how this theory can be viewed as a dimension reduction theory. We discuss the merits of covariate dimension reduction in observational settings and present a theory of Sufficient Dimension Reduction Summaries that encompasses and greatly expands Propensity Theory. This theory identifies new dimension reduction summaries that offer improved estimation relative to use of propensity scores.

A social tea will be held at 3:00 P.M. in A434 Mayo. All are Welcome.
For more details contact 612-624-4655 or see http://www.biostat.umn.edu/seminar_academic.html