Fan's group is interested in statistical methods in financial econometrics and risk managements, computational biology, biostatistics, high-dimensional statistical learning, data-analytic modeling, longitudinal and functional data analysis, nonlinear time series, wavelets and their applications, among others. Our primary research focuses on developing and justifying statistical methods that are used to solve problems from the frontiers of scientific research. This is expanded into other disciplines where the statistics discipline is useful.
In each of the areas mentioned above, our group devotes most of our efforts to the search for intuitively appealing, model-free, robust nonparametric approaches and illustrates the approaches by real data and simulated examples. Modern statistical principles and modeling inevitably involve intensive computation, which is a part of the methodological research development. Our group is also very interested in developing foundational statistical theory and in providing fundamental insights to sophisticated statistical models. These include sampling theory, statisical learning theory, minimax theory, efficient semi-parametric modeling and nonlinear function estimation.
Recently, our group is particularly interested in financial econometrics, risk management, computational biology, biostatistics, high-dimensional data-analytic modeling and inferences, nonlinear time series, analysis of longitudinal and functional data, and other interdisciplinary collaborations.
Students and faculty in the Operations Research and Financial Engineering Department involve in a variety of different projects related to financial economometrics. Financial econometrics is an integration of finance, economics and statistics. It uses statistical techniques and economic theory to address a variety of problems from Finance. These include building financial models, estimation and inferences of financial models, volatility estimation, risk management, testing financial economics theory, capital asset pricing, derivative pricing, portfolio allocation, simulating financial systems, hedging strategies, among others.
Computational biology is a quantitative approach to understand biological and genomic functions and molecular mechanisms. Researchers at the ORFE Department use cutting edge statistical techniques to analyze mRNA and protein expression from microarray and proteomic experiments. Microarray and proteomic technologies have enabled researchers to monitor simultaneously levels of thousands of genes and proteins as they are expressed in tissues, cell lines, patients' specimens, at particular time and under a variety of different conditions. The techniques have been widely used for monitoring mRNA and protein expression in many areas of biomedical research. These kinds of bioinformatic problems are usually large-scale and high experimental variation.
Statistics Theory and Methods
Statistics is a discipline that studies designs, collects and analyzes data from all scientific disciplinary. It pervades every facets of quantitative analysis in science and engineering. The aims of statistics include understanding sampling variability, making inferences and decisions from noisy data, establishing relationship between the covariates and response variables, and predicting future events. The main focus of statistical research at the Department of Operations Research and Financial Engineering is to study the problems arising from Financial Engineering, Security Pricing, Bioinformatics and Health Sciences.