| | Gong Tang, PhD
Assistant Professor, Biostatistics
Statistician, NSABP http://www.pitt.edu/~got1 Education B.S., Mathematics, Beijing University, 1991
M.S., Mathematics, Beijing University, 1994
M.A., Mathematics, Johns Hopkins University, 1996
Ph.D., Biostatistics, University of Michigan, 2001 Research Interests My primary research interest is the analysis of data sets with missing values. Most statistical techniques require complete data. In practice, missing data often occur by design or data attrition and the missingness needs to be justified to avoid invalid inference. Popular approaches, including likelihood-based and estimating-equations- based methods, specify models for the complete data and the missing- data mechanism. I am interested in robust procedures that avoid specification of the mechanism and the corresponding sensitivity analysis. Past works focused on parametric modeling of the complete data. Currently I am working on semi-parametric models for the complete data. Other related works include using auxiliary information to improve efficiency and studying appropriate imputation procedures for datasets with nonignorable nonresponse. Other research interests are longitudinal data analysis and semi-
parametric statistics. As a statistician at NSABP, I have been working on analysis of gene expression data. Collaboration between NSABP and GHI resulted in the OncoType DX assay which is useful for predicting distant recurrence rate and potential benefit from chemotherapy based on the assay results. Courses BIOST 2040: Elements of Stochastic Processes
BIOST 2065: Analysis of Incomplete Data Selected Publications Tang, G., Little, R.J.A., and Raghunathan, T.E. Analysis of Multivariate Missing Data with Nonignorable Nonresponse. Biometrika, Vol. 90, 747-764, 2003. Tang, G., Little, R.J.A., and Raghunathan, T.E. Analysis of Multivariate Monotone Missing Data by A Pseudolikelihood Method. Proceedings of the Second Seattle Symposium in Biostatistics: Analysis of Correlated Data . Lecture Notes in Statistics, Vol. 179, 35-50. Ed. D. Lin and P. J. Heagerty. 2004. New York : Springer-Verlag. Paik, S., Shak, S., Tang, G., Kim, C., Baker, J. Cronin, M., Baehner, R., Walker, M., Waston, D., Park, T., Hiller, W., Fisher, E., Fisher, B., Wickerham, L., Bryant, J. and Wolmark, N. A Multi-Gene RT-PCR Assay Using Fixed, Paraffin-Embedded Tumor Tissue to Predict the Likelihood of Breast Cancer Recurrence in Node Negative, Estrogen Receptor Positive, Tamoxifen-Treated Patients. The New England Journal of Medicine ,2004; 351:2817-26. Contact Information 307 Parran Hall
130 DeSoto Street
Pittsburgh, PA 15261
Telephone: 412-624-3027
Facsimile: 412-624-2183
Email: got1@pitt.edu |  |
