List of researchers in CAMS working on problems related to Statistics: Bhattacharjee, Dhar, Dios, Khan, Jain, Yoo.

Applied Probability and Statistics, as a discipline, is concerned with the study and analysis of processes in which uncertainty plays a significant role. In today's data driven environment in which we live, the need for and utility of uncertainty modeling and statistical analysis is assuming increasing importance in virtually every field of human interest, e.g., in the comparative study of DNA databases, evaluation of drug safety and effectiveness, design and analysis of modern communication protocols, stochastic models in finance, study of aging and performance analysis of components and complex systems, to name a few.

While the field of Applied Probability and Statistics is driven by the need to solve applied problems, its progress and development comes from basic research and from their application to solve specific problems arising in practice. This interplay of basic and applied research has benefited both. Real life applied problems have often posed new theoretical problems which had to be solved by developing new methods (e.g., survival analysis and clinical trials). Conversely, new theoretical ideas and methods which were developed in a specific applied context were later seen to be of much broader applicability to other areas (e.g., nonparametric aging ideas which owe their origins to research in stochastic modeling of hardware reliability of physical systems were later seen as useful constructs in many other areas such as in the studies of queuing systems, stochastic scheduling, branching processes as well as in modeling economic inequality).

The Statistical Consulting Laboratory (SCL), which operates under the umbrella of CAMS, provides methodological / data analysis consulting services to the University community on request, as well as to external clients. Consulting activities channeled through the SCL, are under the overall administrative supervision of a statistics faculty member (currently, A. Jain). Examples of recent consulting projects, in which graduate students were involved to gain valuable hands-on experience, include (i) analysis of demographic and student performance data from public schools to identify student and teacher characteristics that are helpful in predicting student performance, and (ii) survey design to assess the reliability of electronic voting machines.

The current research interests of the Statistics faculty are in the following broad and overlapping areas : distribution theory and statistical inference (Bhattacharjee, Dhar, Khan), minimum distance estimation (Dhar), Bayesian modeling (Bhattacharjee) and Baysian inference (Yoo, Khan), DNA microarray analysis (Khan), orthogonal arrays in experimental designs (Dios), applied probability models (Bhattacharjee, Dhar), statistical theory of reliability and survival analysis (Bhattacharjee), stochastic orders and their applications (Bhattacharjee), discrete multivariate distribution / reliability models and inverse sampling (Dhar), change point problems (Yoo), statistical issues in clinical trials (Dhar), Markov Chain Monte Carlo methods (Yoo), and non-traditional applications of reliability theory (Bhattacharjee).

The rest of this page contains links to examples of the research in Applied Probability and Statistics that have been recently considered by the CAMS members. The links to individual faculty web pages that contain more information can be found at the top of this page.

Bhattacharjee : Integral stochastic orders and comparison of randomly stopped sums (see CAMS Technical Report #35, 2003/04 )

Bhattacharjee: Bayesian modeling of adaptive economic choices (see CAMS Technical Report #36, 2003/04 )

Bhattacharjee: Natural strengthenings of the DFR property and applications (see CAMS Technical Report #18, 2004/05 )

Khan: Bayesian modeling and inference under double censoring (see CAMS Technical Report #13, 2004/05 )

Khan: Predictive inference for future responses (see CAMS Technical Report #14, 2004/05 , CAMS Technical Report #17, 2004/05 )

Yoo: Bayesian hierarchical changepoint models, and their simulation (see CAMS Technical Report #19, 2004/05 )

Yoo: Variable selection in Bayesian hierarchical model for longitudinal biomarkers of prostate cancer (see CAMS Technical Report #35, 2004/05 )