Education / Graduate Education / PhD Programs / Statistics / Curriculum / Historical

Historical Course Information

Graduate Course Schedules

Department of Biostatistics and Computational Biology Course Schedules (BST courses)

University of Rochester Course Schedules (All course offerings, including IND courses)

BST Topic Course Examples (BST 511, BST 512, BST 550, BST 570)

Smoothing Methods
Prerequisites: BST 412 and BST 426
The course will cover 3 main topics: (1) density estimation methods, including histograms, frequency polygons, kernel density estimators, local likelihood density estimators, and penalized likelihood and spline-based estimators; (2) nonparametric regression, with an emphasis on local polynomial modeling and some discussion of cubic smoothing splines; and (3) generalized additive models (GAM). Attention will be paid to locally varying smoothing parameters, boundary effects and corrections, consistency, rates of convergence, extensions to higher dimensions and the "curse of dimensionality," limit distributions, smoothing parameter selection, and kernel choice, including higher-order and equivalent kernels.

Frailty Models in Survival Analysis
Prerequisites: BST 411, BST 479 or BST 514
The notion of frailty provides a convenient way to introduce random effects, association and unobserved heterogeneity into models for survival data. In its simplest form a frailty is an unobserved random proportionality factor which modifies the hazard function of an individual, or of related individuals. In essence the concept goes back to the 1920 work of Greenwood and Yule on "accident proneness." The term frailty itself was introduced by Vaupel, Manton and Stallard in 1979 and applications in survival analysis were popularized in a series of papers by P. Hougaard. Applications to multivariate survival data date from a seminal 1978 paper by D. Clayton. The course will provide a general introduction to frailty models for univariate and multivariate survival data and for repeated events. We will discuss the formulation of frailty models and identifiability aspects, connections with two-sample rank tests for survival data and with measures of association for bivariate survival data. Parametric and semiparametric methods of fitting frailty models will be reviewed, including the use of the EM algorithm and related approaches including hierarchical likelihood. Extensions to correlated frailty models will also be described. The distinction between "conditional models," in which the effect of observed covariates is described conditionally on the value of the random effect and "marginal models," which integrate over the unobserved effect, will be emphasized. Classical survival analysis methodology, including the Kaplan-Meier estimator and Cox's regression model, will be reviewed briefly at the start of the course.

The Bootstrap, The Jackknife, and Resampling Methods
Prerequisites: BST 412
Bootstrap methods, including the nonparametric, parametric, smoothed, m-out-of-n, and randomly weighted bootstrap procedures, will be presented for estimating standard errors, constructing various types of confidence intervals, performing hypothesis tests, and estimating bias. One- and two-sample problems, regression, correlated data, and more general inference problems will be considered. Jackknife and cross-validation methods will also be discussed.

BST Courses No Longer Offered

BST 416 Applied Statistics
Prerequisites: STT 211 or STT 212 or BST 463 or equivalent
One- and two-way analysis of variance; simple and multiple regression; analysis of covariance; analysis of residuals, use of transformations; topics from contingency table analysis and nonparametric statistics. Emphasis on real examples from the biomedical and social sciences, with extensive use of statistical software.

BST 421 Sampling Techniques
Prerequisites: STT 203 or STT 213
Simple random, stratified, systematic, and cluster sampling; estimation of the means, proportions, variance, and ratios of a finite population. Ratio and regression methods of estimation and the use of auxiliary information. The nonresponse problem.

BST 422 Design of Experiments
Prerequisites: BST 416 or BST 464 or BST 476
Basic designs and their principles; randomization; blocking; use of concomitant information.

BST 431 Introduction to SAS
Introduction to SAS programming; statistical analysis in SAS. Topics in statistical analysis provide working examples. Note this course was offered only in Fall 2021; the content has been added to BST 430 Introduction to Statistical Computing.

BST 441 Applied Multivariate Analysis
Prerequisites: BST 476 or BST 426
Methodology and applications of multivariate analysis; Hotelling's T2; multivariate regression and analysis of variance; classification and discrimination; principal components, clustering, and multidimensional scaling; use of statistical software.

BST 451 Exploratory Data Analysis
Prerequisites: BST 416 or BST 430
Graphical techniques to reveal structure in data; model fitting to describe structure; model checking; transformations; outliers and resistant fitting methods.

BST 464 Applied Linear Regression
Prerequisites: BST 463 or equivalent
One-way and two-way analysis of variance; multiple comparisons involving means; fixed and random effects; simple and multiple linear regression; analysis of covariance; interactions; correlation and partial correlation; multicollinearity; model selection; model checking.

BST 466 Categorical Data Analysis
Prerequisites: BST 464 or equivalent
Measures of association for categorical outcomes; contingency table analysis; regression analysis for binary, polytomous, count and time-to-event responses; emphasis on general ideas and applications of models and methods using statistical software such as SAS; review of necessary theory underlying likelihood and nonparametric inference as it pertains to the development of relevant models and test statistics.

BST 476 Introduction to Linear Models
Prerequisites: STT 203 or STT 212 or BST 463
Simple and multiple regression models; least-squares estimation; hypothesis testing; interval estimation; prediction; matrix formulation of the general linear model; polynomial regression; analysis of variance; analysis of covariance; methods for simultaneous inference; residual analysis and checks of model adequacy.