Examines implications for statistical power of using predicted probabilities for a dichotomous variable rather than the actual variable. Information loss from not observing actual values can be large. Direct substitution is easy and efficient.
Examines financial implications of the Centers for Medicare & Medicaid Services Hierarchical Condition Categories risk-adjustment model on Medicare payments for individuals with comorbid chronic conditions. Some payments were underpredicted.
Associate Director, Engineering and Applied Sciences Department; Senior Operations Researcher
Education Ph.D. in industrial and operations engineering, University of Michigan; M.S. in industrial and operations engineering, University of Michigan; B.S. in systems engineering, University of Virginia
Associate Economist; Core Faculty, Pardee RAND Graduate School
Education Ph.D. in economics, University of Texas; M.A. in mathematics, University of Southern California; B.A. in mathematics, Pomona College