Departmental Faculty & Research Profile: 2008
Duke Statistics was created and began operating as an academic center in 1986, and became one of the eight natural science departments in the School of Arts & Sciences in 1991. Duke Statistics activities include campus-wide teaching of entry-level statistics courses for undergraduates and some graduate students, and statistical consulting for the Schools of Arts & Sciences, and Environment and Earth Sciences. The Duke Statistics PhD program, established in 1990, is distinguished by its strong focuses in inter-disciplinary statistics research, Bayesian statistics and scientific computation. Individual faculty members' research interests are described here.Major Research Areas: 2008
Duke Statistics research areas in core statistical science include:
- Bayesian theory, foundations of statistical inference, Bayesian/non-Bayesian interfaces
- Statistical model assessment, selection and combination
- Data mining and machine learning
- Decision theory and risk analysis with applications
- Nonparametric Bayesian methods
- Graphical models
- Monte Carlo simulation: algorithms and theory
- Distributed and cluster statistical computing
- High-dimensional statistical models and large data sets
- Time series analysis and forecasting
- Large-scale statistical mixture modeling
- Sampling, survey methods and missing data imputation
- Spatial statistical modeling and spatial stochastic processes
- Simulation and inference for stochastic processes and random fields
- Extreme values in time series and random fields
- Geometry and topology in statistics; shape analysis
- Complex hierarchical and latent variable modeling
- Multiscale/multiresolution methods
Core research is synergistic with inter-disciplinary applications, key current areas including:
- Computational Biology & Statistical Genetics
- Cancer genetics, risk modeling, genetic epidemiology
- Gene and protein expression analysis and molecular phenotyping
- Gene pathways and networks, especially in cancer biology
- Stochastic modeling in molecular biochemistry and biophysics
- Statistical modeling in systems biology: inference in dynamic stochastic cellular and genetic networks
- Biostatistics and health care policy
- Spatial epidemiology
- Large-scale genetic profiling in clinical studies
- Computer simulation and applied inverse problems
- Signal processing/communications, and network traffic
- High-energy physics, physical chemistry, and materials science
- Ecological forecasting, biodiversity/abundance
- Pollutant modeling and monitoring, spatial statistics, remote sensing
- Atmospheric sciences -- computer model evaluation