Research in Statistics

Research interests of our statistics group include: methodological and theoretical statistics with a strong inter-disciplinary research component with applications in biomedical areas, environmental science, climate, public health, and epidemiology. Statistics faculty work in diverse areas with interests in Bayesian statistics, statistical computing, spatial and spatio-temporal statistics, hierarchical modeling, catagorical data analysis, genomics and proteomics, objective Bayesian methods, Bayesian robustness, subgroup analysis, nonparametric, non-linear modeling, bioinformatics, epidemiology, mixed models of uncertainty, fuzzy sets and their applications, statistical decision theory (in general and in the presence of fuzziness), testing of statistical hypotheses, model selection in general and using set-valued data.

Statistical faculty have joint research collaborations with several local, national, and international institutions. They help graduate students with finding valuable paid internships in statistics at local industries and research institutions while at UC and with jobs after graduation. The inter-disciplinary research of our statistics faculty and our statistics graduate program are funded by several agencies and institutions, including NIH, DARPA, JPL, and CCHMC. Our graduate program has a history of extremely successful job placement for our students in industry and academia.


  • Cai, Biao: His research primarily focuses on statistical machine learning with complex data structures. Methodologically and theoretically, his work encompasses areas such as graph learning from multivariate point process data, tensor completion, tensor clustering, tensor regression, and multiomics data analysis. In terms of applications, he has been involved in projects related to learning firm-level networks and social media user behaviors, clustering of medical images, and RNA sequencing data analysis.
  • Cao, Xuan: Bayesian model selection in high-dimensional settings, high-dimensional inference on graphical models with application to genomics and eQTL analysis, and statistical applications in neuroscience.
  • Chang, Won: Analysis of high-dimensional and non-Gaussian spatial data; Computer model emulation and calibration using deep neural networks and Gaussian processes; Missing value imputation and survey data analysis using deep neural network, Bayesian analysis, and pseudo likelihood; Spatial clustering; Statistical and machine learning methods for atmospheric science, glaciology, hydrology, genetics, epidemiology, and business analytics.
  • Kang, Emily: methodology in spatial and spatio-temporal statistics, hierarchical statistical modeling in environmental and climate services, Bayesian and empirical Bayesian methods for large or even massive spatial and spatio-temporal datasets, applications in remote-sensing, uncertainty qualification for climate models, climate sensitivity detection through fluctuation-dissipation theory.
  • Kim, Hang Joon: Bayesian methodology for survey data; missing (or faulty) data problem; Bayesian computing; statistical disclosure limitation; statistical genomics and genetics.
  • Konomi, Alex: Bayesian statistics, methodology in spatial and spatio-temporal statistics, uncertainty quantification, computational methods of large or even massive spatial data sets and computer simulations ("big data"), Bayesian trees for non-parametric regression, computer experiments, computer model calibration, sequential design, image analysis, variable selection (big "p" small "n").
  • Ralescu, Dan: Main areas of research: probability and statistics with inexact data, mixed models of uncertainty, fuzzy sets and systems and their applications. He has co-authored the first comprehensive book on fuzzy systems and applications, as well as he pioneered limit theorems for random sets in Banach space. He is the associate editor and editor of several major journals on fuzzy systems and intelligent informatics. His theoretical research interests include: random sets, Choquet and other non-linear integrals, non-additive set-functions, mixture models of statistical and non-statistical uncertainty, and non-linear admissible estimators. His interests in applications are in : minimum contrast estimation for random sets, testing of hypotheses with inexact data, hierarchical models for normal random sets, fuzzy control systems, interval-valued probabilities, Bayesian inference with fuzzy prior information.
  • Sivaganesan, Siva: Research is mainly in the general area of Bayesian analysis. His theoretical and methodological research areas include Robust and Objective Bayesian analyses, Subgroup Analysis, Bayesian nonparametrics, Spatial statistics, Survival analysis, Errors in variables, and nonlinear modeling, His research is geared mainly towards applications in the general area of biomedical research, with particular focus on Bioinformatics, Clinical trials, PK/PD, and Epidemiology. 
  • Song, Seongho: works in the general area of developing Bayesian statistical models including Hierarchical Bayesian Inference with MCMC methods. His particular areas of interest are related to (1) Population Genetics, (2) Lifetime data analysis with various type of censoring methods, (3) Econometrics with stochastic frontier models, discrete choice models, (4) Longitudinal Data Analysis and (5) Microarray data analysis. Recently I have started working on the Uncertainty Quantification for the climate control data.
  • Wang, Xia: Bayesian methodology and computation, categorical data analysis, spatial and spatio-temporal statistics, funtional data analysis, applications in social sciences, proteomics, bioinformatics, ecological and environmental sciences, and climate change assessments.