Principles of data analysis and advanced statistical modeling. Bayesian inference, prior and posterior distributions, multi-level models, model checking and selection, stochastic simulation by Markov Chain Monte Carlo. Not open to students with credit for STA602. Prerequisites: MATH202, MATH216, MATH216D or MATH221, and STA210, STA230, STA250D. One course.
Quantitative methods for decision making under uncertainty. Probability theory, personal probabilities and utilities, decision trees, ROC curves, sensitivity analysis, dominant strategies, Bayesian networks and influence diagrams, Markov models and time discounting, cost-effectiveness analysis, multi-agent decision making, game theory. Prerequisite: STA230. One course.
A practical introduction to statistical programming focusing on the R programming language. Students will engage with the programming challenges inherent in the various stages of modern statistical analyses including everything from data collection/aggregation/cleaning to visualization and exploratory analysis to statistical model building and evaluation. This course places an emphasis on modern approaches/best practices for programming including: source control, collaborative coding, literate and reproducible programming, and distributed and multicore computing.
Design and analysis of surveys, including random sampling, stratification, clustering, and multi-stage sampling. Design-based and model-based inference. Methods for handling missing data. Prerequisites: STA210 or ECON208D. One course.
Design of randomized experiments and observational studies. Role of randomization, block designs, factorial designs, fractional factorial designs, matching. Analysis of variance, contrasts, propensity score matching, instrumental variables. Prerequisites: STA210, STA250/MATH342. One course.
An introduction to the concepts, theory, and application of statistical inference, including the structure of statistical problems, probability modeling, data analysis and statistical computing, and linear regression. Inference from the viewpoint of Bayesian statistics, with some discussion of sampling theory methods and comparative inference. Applications to problems in various fields. Prerequisites: MATH202, MATH212 or MATH222, and STA230 or MATH340. One course.
Extensive study of regression modeling. Multiple regression, weighted least squares, logistic regression, log-linear models, analysis of variance, model diagnostics and selection. Emphasis on applications. Examples drawn from a variety of fields. Prerequisite: Statistics 100-level course. Permission of Director of Undergraduate Studies required for courses outside Statistical Science. One course.
Introduction to probability, independence, conditional independence, and Bayes’ theorem. Discrete and continuous, univariate and multivariate distributions. Linear and nonlinear transformations of random variables. Classical and Bayesian inference, decision theory, and comparison of hypotheses. Experimental design, statistical quality control, and other applications in engineering. Not open to students who have taken STA250 or STA611. Prerequisite: MATH212 or equivalent. One course.