George T. Duncan

I like to use the ideas of statistical decision theory to tackle complicated issues of public concern, like the environment, health management, and dispute resolution. Right now, I am deeply involved in issues of resolving the tension between maintaining privacy and providing information. More generally, I teach and do research on social accountability aspects of information technology, with a focus on the World Wide Web. An intriguing activity for me was chairing the National Academy of Sciences Panel on Confidentiality and Data Access, which included economists, sociologists, psychologists, computer scientists, and lawyers. Now I chair the American Statistical Association's Privacy and Confidentiality Committee, and so get to interact with statisticians from government, business, and industry. What fascinates me about my research on confidentiality in statistical databases is how the mathematical ideas of Bayesian analysis, optimization algorithms, and linear algebra can be used to limit disclosure of sensitive information. With my appointment in the Heinz School of Public Policy and Management, I work on the interplay of public policy problems and statistical models. I much enjoy collaboration with students as they discover how statistics can add so much insight to problems of public interest.

Some Related Publications

Duncan, G. T. and Kaufman, S. (1996). Who should manage information and privacy conflicts? Institutional design for third-party mechanisms. International Journal for Conflict Management, 7, 21-44.

Duncan, G. T., Krishnan, R., and Mukherjee, S. (1995). Inference channel detection in multilevel relational databases: A graph-based approach. Journal of Organizational Computing, 5, 123-138.

Duncan, G. T., Jabine, T., and de Wolf, V. (1993). Private Lives and Public Policies. Washington, D. C.: National Academy Press, 274 pages.

Duncan, G.T., Gorr, W. and Szczypula, J. (1993). Bayesian forecasting for seemingly unrelated time series data: application to local government forecasting. Management Science, 39, pp. 275-293.