An introduction to modern-day extensions of simple linear regression and ANOVA to the chi-square test including logistic regression and log-linear modelling techniques based on generalized linear models. Methods for matched-pair, small datasets, ordinal and multi-category data also discussed. This course focuses on applications using real-life data sets, and uses popular software packages.

This course will cover the basic principles of nonparametric methods in statistics including: one, two and K sample location methods; tests of randomness; tests of goodness of fit; nonlinear correlation; histogram; density estimation; nonparametric regression. Students should learn how to apply the nonparametric techniques in real datasets, interpret the results and draw conclusions.

While simultaneously reviewing basic statistical methods (t tests, regression, ANOVA, interaction, etc.), this course introduces statistical modelling using the SAS program, involving the DATA step and various SAS procedures. Working on hands-on projects using real datasets, students present their final project results.

As the first part in a two-semester sequence, this course introduces basic principles of probability including combinatorial methods, probability and cumulative density and mass functions, moment generating functions and applications, expected values and variance and other moments, and order statistics. This course emphasizes related theorems and proofs.

As the second part in a two-semester sequence, this course thoroughly explores the central limit theorem and its variants and uses, estimation, hypothesis testing, sufficiency, efficiency, uniformly most powerful methods, information, and asymptotic methods. Time permitting, Bayesian topics may also be explored and discussed.

This course addresses topics such as finite-state Markov processes and Markov chains, classification of states, long-run behavior, continuous time processes, birth and death processes, random walks, and Brownian motion.

This course provides students with a thorough introduction to statistical experimental design and to the statistical methods used to analyze the resulting data. The concepts of comparative experiments, ANOVA and mean separation procedures will be reviewed; blocking (complete and incomplete) will be discussed, as will be factorial designs, fractional factorial designs, and confounding. The course will focus on biometric applications such as clinical trials, HIV studies, and environmental and agricultural research, but industrial and other examples will occasionally be provided to show the breadth of application of experimental design ideas.

This course provides students with a thorough introduction to applied regression methodology. The concept of simple linear regression will be reviewed and discussed using matrices, and multiple linear regression, transformations, diagnostics, polynomial regression, indicator variables, model building and multicolinearity will be discussed, as will be nonlinear and generalized linear regression. The course will focus on applications such as those from biometry and biostatistics (clinical trials, HIV studies, etc.), sports, engineering, agriculture and environmental science.

This course provides an introduction to modern-day extensions of simple linear regression and ANOVA to the chi-square test including logistic regression and log-linear modeling techniques based on generalized linear models. Specialized methods for ordinal data, small samples, multi-category data, and matched pairs will also be discussed. The focus throughout this course will be on applications and real-life data sets.

Modern statistical methods are covered to analyze data that is right, left and/or interval-censored. Nonparametric approaches such as the Kaplan-Meier estimation technique, log-rank test and proportional-hazards model are considered as are parametric methods such as those based on the Exponential and Weibull distribution. Accelerated failure time models and nonlinear models are also discussed.

This course uses SAS, R and high-level languages to perform statistical modelling by conducting statistical simulations to assess linear, generalized linear, nonlinear and complex models and experimental designs. Students will gain practical experience and knowledge in real-world statistical situations for which underlying theory is cumbersome or otherwise intractable.

This course presents the role of likelihood methods in a whole range of statistical problems. The course reviews theoretical developments such as efficiency, completeness, and the Cramer-Rao lower bound, and shows how the likelihood approach is used to surpass these methods and to analyze regression problems, to deal with nuisance parameters by using marginal likelihood methods, and to deal with complex data structures such as censored and spatial data.

This course covers experimental design (including interaction, analysis of covariance, and crossover designs) and the analysis of designed studies, simple and multiple linear regression, generalized linear and nonlinear regression, bioassay, relative potency and drug synergy, multivariate analysis (including MANOVA and multivariate regression), repeated measures (designs and analysis), and survival analysis (Cox proportional odds, log-rank tests, Kaplan-Meier estimation) of censored data. The emphasis of the course will be on applications instead of statistical theory, and students will be required to analyze real-life datasets using popular statistical packages.

This course explores recently developed mathematical, probabilistic and statistical methods currently used in the fields of bioinformatics and DNA microarray and protein array data analysis. These include stochastic processes, (hidden and traditional) Markov chains, tree- and clustering techniques (including principal components analysis and biplots), discriminant analysis, experimental design strategies and ANOVA methods. Our focus in this course is on the application of these techniques and on meaningful interpretation of results.

This course provides an introduction to the models and methods used in actuarial mathematics and risk theory. Students are expected to gain a broad understanding of frequency and severity modelling, pricing, and accumulated risk. This course includes a blend of theory and applications.

With an introductory background in the field provided in STAT 465, this course thoroughly explores modelling and estimation techniques in actuarial mathematics and risk theory.

With a focus on insurance, pensions and investments, this course provides an overview of the theory of risk, emphasizing the statistical challenges and assumptions inherent in models and methods.

This topics course provides the means for new courses on current or 'hot' topics to be offered to students, with the topics being crafted to the given topic at hand. As such, the course may be taken repeatedly.

Working with a statistics faculty member on a one-on-one or small group format, this course affords students the opportunity to thorough explore a statistical topic at greater depth. Generally, it involves a good deal of outside reading and/or programming, and weekly meetings with the professor.