Assistant Professor Elizabeth Malloy

Education       

PhD, Statistics, University of New Mexico
MA, Mathematics, University of New Mexico
BS, Electrical Engineering, Virginia Polytechnic Institute and State University

I received my PhD in Statistics from the University of New Mexico, where I taught Mathematics for Elementary School Teachers, College Algebra, Calculus, Introduction to Statistics, and the Elements of Probability and Statistics.  I am currently teaching Basic Statistics and Regression at American University.  My research interests include functional data analysis, biostatistics, semiparametric regression, and computational statistics, with applications in environmental statistics, occupational health, psychology, and other areas.  As a Research Fellow at the Harvard School of Public Health, I developed methodology that incorporates functional predictors into linear mixed models.  This problem was motivated by a study of the short-term effects of exposure to traffic pollution on lung function in susceptible populations where exposures were recorded at regular time intervals.  I am also interested in semiparametric regression, particularly in survival models.  I have investigated the relationship between rectal cancer mortality and cumulative exposure to metalworking fluids in a study of an autoworkers cohort.  I am currently examining properties of model fit criteria and comparing the use of penalized splines to more traditional methods in Cox regression models using Monte Carlo simulations.

Recent presentations:

• How well do penalized splines fit exposure-response curves in environmental epidemiology?  A simulation study of six typical scenarios, poster, Department of Environmental Health Symposium, Harvard School of Public Health, April 2006
• Wavelet-based functional mixed models, journal discussion leader, Functional Data Analysis Working Group, Department of Biostatistics, Harvard School of Public Health, April 2006
• Bayesian smoothing in functional linear mixed models using wavelet shrinkage, invited, ENAR, March 2006
• Bayesian smoothing in functional linear mixed models using wavelet shrinkage, Environmental Statistics Seminar Series, Department of Biostatistics, Harvard School of Public Health, March 2006
• Examining model fit for penalized splines in Cox models, Department of Environmental Health, Harvard School of Public Health, January 2006
• An introduction to R programming, Department of Environmental Health, Harvard School of Public Health, November 2005
• Generalized linear models with functional predictors, journal discussion leader, Functional Data Analysis Working Group, Department of Biostatistics, Harvard School of Public Health, November 2005
• An introduction to using MATLAB in environmental statistics, Environmental Statistics Working Group, Department of Biostatistics, Harvard School of Public Health, October 2005
• Introduction to functional data analysis, Functional Data Analysis Working Group, Department of Biostatistics, Harvard School of Public Health, September 2005
• Model fit criteria in Cox models, contributed, ENAR, March 2005
• Diagnostics for the scale of functional predictors, Environmental Statistics Seminar Series, Department of Biostatistics, Harvard School of Public Health, January 2005
• Diagnostics for functional predictors in logistic regression, poster, University of Florida Fifth Annual Winter Workshop: An IMS Mini-Meeting on Functional Data Analysis, January 2003

Recent articles include:

• Malloy, E.J., Miller, K.L. and Eisen, E.A. Rectal cancer and exposure to metalworking fluids  in the automobile industry" Occupational and Environmental Medicine (in press)
• Malloy, E.J., Bedrick, E.J. and Goldsmith, T. "Diagnostics for the scale of functional predictors in generalized linear models" (in revision for Technometrics).
• Miller, K.L., Malloy, E.J. and Eisen, E.A. "Healthy worker survivor effect in the Vermont granite workers study" (in preparation).
• Malloy, E.J., Coull, B.A., Morris, J.S., Dubowsky, S.D. and Suh, H.H. "Bayesian smoothing of functional predictors in linear mixed models using wavelet shrinkage " (in preparation).
• Malloy, E.J., Spiegleman, D., and Eisen, E.A. "Penalized spline smoothing parameter selection in Cox models: a simulation study" (in preparation).