Matrix algebra, Multivariate normal distributions, discriminant analysis, canonical correlations, and Multivariate analysis of variance.

This course will cover the topics in the theory and practice of Bayesian statistical inference, ranging from a review of fundamentals to questions of current research interest. Motivation for the Bayesian approach. Bayesian computation, Monte Carlo methods, asymptotics. Model checking and comparison. A selection of examples and issues in modeling and data analysis. Discussion of advantages and difficulties of the Bayesian approach. This course will be computationally intensive through analysis of data sets using the R statistical computing language.

Topics included are survival function, hazard function, right censoring, nonparametric methods for comparing two survival distributions, parametric and nonparametric regression methods with survival data.

Topics included are clinical trials, longitudinal data analysis, Bayesian method, and diagnosis.

Functional data arises in many scientific areas. With an increasing number of cases, the collected data are curves or functions. This course introduces basic concepts and methods of functional data analysis. Topics include: functional data representation, smoothing methods, exploration of functional data, registration of functional data, functional principal component analysis, canonical correlation and discriminant analysis, functional linear models.

Topics included are statistical inference, Multivariate normal distribution, distribution of quadratic forms, linear models, regression models and experimental design models.

Topics included are statistical inference, Multivariate normal distribution, distribution of quadratic forms, linear models, regression models and experimental design models.

The course is primarily on the theory of statistical estimation. Topics included are method of moments, least squares, maximum likelihood, minimum variance unbiased estimation, the sufficiency and completeness of statistics, Fisher information and Cramer-Rao bounds, statistical decision theory, and Bayes decision rules.

Classical and modern statistics, probability, decision theory, estimation theory, testing hypotheses, confidence intervals, large sample theory, and sequential analysis.

Random variables and expectations, distribution and characteristic functions, laws of large numbers and central limit theorem, conditional probability, and expectation.

Introduction to stationary stochastic processes, spectral representations; Box-Jenkins time series models; forecasting methods. Applications include use of a statistical computer package.

Analysis of randomized and incomplete block designs; factorial and nested designs using fixed, random, and mixed effects models. Applications include use of a statistical computer package.

Topics included are optimization, numerical integration, bootstrapping, cross-validation and Jackknife, density estimation, smoothing, and use of the statistical computer package of S-plus/R.

Topics included are the Monte Carlo method for integration, Metropolis-Hastings algorithms, the Gibbs sampler and other Markov chain-based methods, importance sampling, simulated tempering, perfect sampling, and other related subjects. Some applications will be illustrated by real examples. Applications include use of a statistical computer package.

This course covers programming using the SAS statistical software package, and it provides an introduction to data analysis stressing the implementation using SAS. Topics include two main parts: 1) SAS programming: data management and manipulation, basic procedures, macro programming; 2) data analysis: descriptive statistical analysis, one- and two-sample inference, basic categorical data analysis, regression analysis, and other selected topics.

Three lecture hours a week. Nonparametric testing and estimation procedures are introduced. Topics include rank methods for one sample and two sample problems, rank tests for one-way layouts, linear regression and independence problems, robust estimates, goodness-of-fit test, U-statistics, recent developments in nonparametric statistics.

May be repeated for credit if topics vary.

May be repeated for credit.

May be repeated for credit.

This course involves supervised, off-campus experience in a field in which statistical science plays an important role.

This course involves opportunity to study new research topics. Students will develop advanced statistical skills in the frontier of biostatistics.

Analysis of Multinomial Data and Contingency tables, loglinear model for count data, model selection procedures; applications include use of statistical software packages, like SAS and S+.

Sampling from finite populations; random, stratified, cluster, and systematic sampling; estimation of means and variances; and ratio and regression sampling.

The objectives of Statistical Consulting Practicum are for students to develop skills needed by a successful statistical consultant, such as the effective collaboration skills with researchers in other disciplines, the communication skills, both written and verbal. In the orientation phase, a number of lectures will cover topics on consulting philosophy, effective problem identification, consulting session management and positive interaction with clients. In the practicum phase, students will be exposed to realistic statistical and scientific problems.

The objectives of Statistical Consulting Practicum are for students to develop skills needed by a successful statistical consultant, such as the effective collaboration skills with researchers in other disciplines, the communication skills, both written and verbal. In the practicum phase, students will be exposed to more realistic statistical and scientific problems.

Directed research leading to a research paper in statistics or analysis of a statistical problem. This course is intended to satisfy the requirement for a research paper or a written report of a laboratory experience for the non-thesis option.

The course is to serve the need of graduate students who are ready to start research work. It provides students the opportunity to see a wide range of topics that are currently being studied by statisticians. It also helps students to learn the important elements of a successful professional talk and to develop skills of professional communication and presentation.

octoral Dissertation Research.