Professor
Johns Hopkins University
Bloomberg School of Public Health
Department of Epidemiology
For additional information click here .
|
Christopher Cox, Ph.D.
Professor Johns Hopkins University Bloomberg School of Public Health Department of Epidemiology |
| Dr. Christopher Cox joined the Statistics in Epidemiology (STATEPI) team in February 2005. He holds a Ph.D. degree in mathematics from the University of Illinois (1973). Before coming to the Department of Epidemiology of the Johns Hopkins Bloomberg School of Public Health, Dr. Cox was a Senior Research Fellow in the Division of Epidemiology, Statistics and Prevention Research (DESPR) at NICHD, beginning in 2002. His previous long academic career at the University of Rochester Medical Center (1975-2001) moved from Assistant Professor to Full Professor in the Department of Biostatistics, with joint appointments in the Departments of Environmental Medicine and Psychiatry. His principal research interest is in the efficient computation of maximum likelihood estimates for single parameter exponential family and more general nonlinear regression models (generalized nonlinear models), and in the practical application of these models in biomedical research. An important aspect of this research is the application of the delta method to provide standard errors for nonlinear functions of model parameters. For example, a recent paper employed the delta method to provide a standard error for a new definition of the limit of quantitation for a laboratory assay in which actual concentrations are estimated from measured values by means of a linear calibration curve. A second paper involved an application of the delta method to provide a simplified computation of the standard error for a model-based estimate of the attributable risk under both case-control and cross-sectional sampling. Since coming to STATEPI, Dr. Cox has become involved in several projects. He is collaborating with Dr. Muņoz and Dr. Chu on a study of the generalized gamma family of parametric survival models. This family includes all of the common types of hazard under one umbrella, and provides a useful alternative to semi-parametric approaches that require the stringent assumption of proportional hazards. Graphical tools for the presentation of relative times and relative hazards are an important aspect of this work. Variability of these estimates is examined using the delta method and bootstrap resampling techniques. Dr. Cox is a member of the Behavioral Research Working Group and is involved in a number of collaborative projects in both the MACS and WIHS. For additional information click here . |
|
Click here to return to the STATEPI Faculty & Staff list.
To see a current list of study publications click here
for a searchable database.
This page was updated January 2006