This course will provide non-biostatistics students seeking a Graduate Biomedical Sciences (GBS) degree with the ability to understand introductory biostatistics concepts. 3 hours. As needed.

Students will gain a thorough understanding of basic analysis methods, elementary concepts, statistical models and applications of probability, commonly used sampling distributions, parametric and non-parametric one and two sample tests, confidence intervals, applications of analysis of two-way contingency table data, simple linear regression, and simple analysis of variance. Students are taught to conduct the relevant analysis using current software such as the Statistical Analysis System (SAS). 3 hours. Fall/Spring.

This course will introduce students to the basic principle of tools of simple and multiple regression. A major goal is to establish a firm foundation in the discipline upon which the applications of statistical and epidemiologic inference will be built. Prerequisite: BST 611 or Permission of Instructor. 3 hours. Spring/Summer.

Continuation of concepts in BST 611/612, intended to introduce students to additional general concepts in biostatistics beyond an introductory level. The course will include a broad overview of three areas:

1. categorical, ordinal, and count methods with proportional odds model and Poisson regression;
2. survival analysis and event outcome data with Kaplan-Meier, proportional hazards, and repeated events;
3. repeated measures, mixed models, hierarchical modeling for longitudinal and missing data.

Study design, analysis interpretation of results, power and sample size estimation, and non-parametric alternatives will be presented for all topic areas. Prerequisite: BST 601 or 611 and 612. 3 hours. Fall.

Vector and matrix definitions and fundamental concepts; matrix factorization and application. Eigenvalues and eigenvectors, functions of matrices, singular and ill-conditioned problems. Prerequisites: BST 622. 3 hours. As needed.

Mathematically rigorous coverage of applications of statistical techniques designed for Biostatistics majors and others with sufficient mathematical background. Statistical models and applications of probability; commonly used sampling distributions; parametric and nonparametric one and two-sample tests and confidence intervals; analysis of contingency tables; simple linear regression and analysis of variance. Prerequisites: A year of calculus and linear algebra. 3 hours. Fall.

Continuation of concepts in BST 621, extended to multiple linear regression; analysis of variance, analysis of covariance, multiple analysis of variance; use of contrasts and multiple comparisons procedures; simple and multiple logistic regression, and an introduction to survival analysis. Prerequisites: BST 621 with a grade of B or higher. 3 hours. Spring.

Simple and multiple regression using matrix approach; weighted and nonlinear regression; variable selection methods; modeling techniques; regression diagnostics and model validation; systems of linear equations; factorial designs; blocking; an introduction to repeated measures designs; coding schemes. Prerequisite: BST 622 with a grade of B or higher. 3 hours. Fall.

Intermediate experimental design and analysis of variance models using matrix approach. Factorial and nested (hierarchical) designs; blocking; repeated measures designs; Latin squares; incomplete block designs; fractional factorials; confounding. Prerequisites: BST 621 and 622 or BST 611 and 612. 3 hours. As needed.

Concepts of clinical trials; purpose, design, implementation, and evaluation. Examples and controversies presented. Prerequisite: BST 611 and 612 or permission of the instructor. Pass/No Pass. 3 hours. Summer.

A hands-on exposure to data management and report generation with one of the most popular statistical software packages. 3 hours. Fall.

This course is restricted to MSPH and DrPH students. This course is an introduction to probability concepts and statistical inference. Topics include counting techniques, discrete and continuous univariate and multivariate random variables & common distributions, probability, expectation, variance, confidence intervals, the Central Limit Theorem, and hypothesis testing. Prerequisite: Calculus II. 3 hours. Spring (may be offered in alternate years depending upon course demand).

Fundamentals of probability; conditional probability and independence; distribution, density, and mass functions; random variables; moments and moment generating functions; discrete and continuous distributions; exponential families, joint, marginal, and conditional distributions; transformation and change of variables; convergence concepts; sampling distributions; order statistics; random number generation. Prerequisite: Advanced calculus. 4 hours. Fall.

Point interval estimation; sufficiency and completeness; ancillary statistics; maximum likelihood and moment estimators; best-unbiased estimator; hypothesis and significance testing; likelihood ratio tests and uniformly most powerful tests; confidence interval estimation; asymptotic properties of estimators and tests; introduction to Bayesian inference. Prerequisite: BST 631 with a grade of B or higher. 4 hours. Spring.

Properties of statistical tests; order statistics and theory of extremes; median tests; goodness of fit; tests based on ranks; location and scale parameter estimation; confidence intervals; association analysis; power and efficiency. Prerequisite: BST 622, BST 632. 3 hours. As needed.

Intermediate level course with emphasis on understanding the discrete probability distributions and the correct application of methods to analyze data generated by discrete probability distributions. The course covers contingency tables, Mantel-Haenszel test, measures of association and of agreement, logistic regression models, regression diagnostics, proportional odds, ordinal and polytomous logistic regression, Poisson regression, log-linear models, analysis of matched pairs, and repeated categorical data. Prerequisite: BST 622 or equivalent recommended. 3 hours. Fall.

Analysis and interpretation of multivariate general linear models including multivariate regression, multivariate analysis of variance/covariance, discriminant analysis, multivariate analysis of repeated measures, canonical correlation, and longitudinal data analysis for general and generalized linear models. Extensive use of SAS, SPSS, and other statistical software. Prerequisite: BST 623. 3 hours. As needed.

Basic principles of measurements; factor analysis and latent variable models; multivariate predictive models including mediation mechanisms and moderator effects; path analysis; integrative multivariate covariance models, methods of longitudinal analysis. Prerequisite: BST 623. 3 hours. As needed.

Kaplan-Meier estimation; Parametric survival models; Cox proportional hazards regression models; sample size calculation for survival models; competing risks models; multiple events models. Prerequisite: BST 622. 3 hours. (Spring)

Simple random, stratified, cluster, ratio regression and systematic sampling; sampling with equal or unequal probabilities of selection; optimization; properties of estimators; non-sampling errors; sampling schemes used in population research; methods of implementation and analyses associated with various schemes. Prerequisite: BST 631. 3 hours. As needed.

Statistical methods and inference through metal analysis. Prerequisites: BST 622 and 623. 3 hours. As needed.

This class will introduce students to population genetics, genetic epidemiology, microarray and proteomics analysis, Mendelian laws, inheritance, heritability, test cross-linkage analysis, QTL analysis, human linkage and human association methods for discrete and quantitative traits. Prerequisite: BST 611 or BST 621. 3 hours. Spring (odd years).

The purpose of this class will be to teach graduate students statistics methods that underlie the analysis of data generated by high throughput genomic technologies, as well as issues in the experimental design and implementation of these technologies. High throughput technologies that will be covered include microarrays, proteomics, and second-generation sequencing. Prerequisites: BST 611 or 621. BST 675 recommended. 3 hours. Spring (even years).

This course is mainly focused on R and how to use R to conduct basic statistical computing. The course contains three themes: R programming, introduction to high performance computing, and basics of statistical computing. Prerequisites: BST 621, BST 622, and BST 626 (Introductory Probability and Inference) or equivalent. 3 hours. (Summer)

Acquire skills for teaching in higher education, including syllabus design, communication skills for the classroom and office hours, creating assignments and rubrics, preparing and giving lectures, preparing nondidactic content, and effective grading. Prerequisites: Must have completed the course that you will be the TA, or similar course, in a prior semester with a grade of B or higher. Completed the Biostatistics Qualifying Exam at the applicable level, have an overall GPA of 3.0 or higher (be a student in good standing with the UAB Graduate School). Receive an invitation from the applicable faculty member to register for this course. 3 hours.

Students will work individually to address, analyze and present the results of an applied problem or grant design each week. The presentation of approaches, solutions and designs will be conducted in a round table format. Students will be evaluated on the quality of solution and by their presentation and class participation. Prerequisites: BST 621 and BST 622. 3 hours.

This course provides an opportunity for students to learn about ongoing research in the field of biostatistics, clinical trials, and statistical genetics. Reserved for BST students. Pass/No Pass. 1 hour. Fall/Spring/Summer.

This course is designed to cover special topics in Biostatistics that are not covered in regular 600 level courses, but suited for Masters students in Biostatistics and doctoral students in other related disciplines. 1-3 hours.

Field experience under joint direction of appropriate public health faculty member and qualified specialists working in selected aspects of public health.Prerequisites:BST 601 or BST 611 and BST 612, ENH 600, EPI 600, HB 600, and HCO 600. Pass/No Pass. 3 hours.

Independent non-thesis research with guidance of appropriate faculty. Pass/No Pass. 1-12 hours.

Multivariate normal distributions and quadratic forms; least square estimation; nested models; weighted least squares, testing contrasts; multiple comparisons; polynomial regression; maximum likelihood theory of log-linear models. Prerequisite: BST 632. 3 hours. Fall (odd years).

Intermediate experimental design and analysis of variance models using matrix approach. Factorial and nested (hierarchical) designs; blocking; repeated measures designs; Latin squares; incomplete block designs; fractional factorials; confounding. Prerequisites: BST 621, 622 (or BST 611/612), 623, 631 and 632. 3 hours. As needed.

This course will provide students with a basic understanding of the fundamental statistical principles involved in the design and conduct of clinical trials. Important topics of discussion will include data management, quality assurance, endpoints, power analysis, interim analysis, adaptive designs, and genetic issues in clinical trials. Prerequisites: BST 611, 612 or 621, 622 and 625. 3 hours. Spring (odd years).

This course builds on the knowledge gained in BST 725 in order to develop a more thorough understanding of the basic methodology behind power analysis, interim data monitoring, and analysis of missing data. The class will involve discussions of recent publications dealing with current topics of interest in clinical trials. The course is offered in 4 stand-alone models: Power and Sample Size Estimation, Interim Data Analysis, Advanced Trial Design, or Complex Outcome Analysis. Prerequisites: BST 621, 622, 625, 630 or (631 and 632) and 725. 1-3 hours. Offered at discretion of faculty.

Stochastic convergence and fundamental inequalities; weak convergence and the central limit theorems; large sample behavior of the empirical distribution and order statistics; asymptotic behavior of estimators and tests with particular attention to LR, score, and Wald tests. Prerequisites: BST 631 and 632. 4 hours. Spring (odd years).

To introduce the student to the basic principles and tools of Bayesian Statistics and most importantly to Bayesian data analysis techniques. A major goal is to establish a firm foundation in the discipline upon which the applications of statistical and epidemiologic inference will be built. The practical part of the course will be based on Bugs (either WinBugs or OpenBugs), possibly accessed through R with the existing tools for the interface (R packages: R2WinBugs or BRugs, coda). This will enable participants to take the practical examples all the way to the reporting stage in terms of tabulations and graphics. Prerequisites: BST 632. 3 hours. Fall (even years).

To illustrate advanced approaches to Bayesian modeling and computation in statistics. We begin with a brief description of the basic principle and concepts of Bayesian statistics. We then study advanced tools in Bayesian modeling and computation. A variety of models are covered, including multilevel/hierarchical linear and generalized linear models, models for robust inference, mixture models, multivariate models, nonlinear models, missing data, and Bayesian model selection. We also introduce some applied areas of modern Bayesian methods, such as genetics/genomics and clinical trials.

The practical part of the course will be based on Bugs (either WinBugs or OpenBugs), possibly accessed through R with the existing tools for the interface (R packages: R2WinBUGS or BRugs, coda). This will enable participants to take practical examples all the way to the reporting stage in terms of tabulations, graphics, etc. Prerequisites: BST 631 and 632. BST 740 would be helpful but not absolutely required. 3 hours. Fall (odd years).

Poisson processes; random walks; simple diffusion and branching processes; recurrent events; Markov chains in discrete and continuous time; birth and death process; queuing systems; applications to survival and other biomedical models. Prerequisite: BST 632. 3 hours. As needed.

Generalized linear models; mixed models; and generalized estimating equations. Prerequisite: BST 723. 3 hours. Spring (even years).

Numerical algorithms useful in biostatistics including likelihood maximization using the Newton-Raphson method, EM algorithm, numerical integration using quadratic and Monte-Carlo methods, interpolation using splines, random variate generation methods, data augmentation algorithm, and MCMC and Metropolis-Hastings algorithm; randomization tests; resampling plans including bootstrap and jackknife. Prerequisites: BST 632. 3 hours. Fall (even years).

This course will provide a statistical basis for describing variation in qualitative (disease) and quantitative traits. This will include decomposition of trait variation into components representing genes, environment and gene-environment interaction. Resemblance between relatives and heritability will be described. Important topics of discussion will include oligogenic and polygenic traits, complex segregations analysis, methods of mapping and characterizing simple and complex trait loci. Prerequisites: BST 623, BST 632, and BST 675. It is assumed that students are comfortable with regression theory, covariance, correlation, and likelihood theory. Interested students are urged to contact the instructors with concerns regarding assumed knowledge. 3 hours. Fall (odd years).

This course builds on the knowledge gained in BST 775 with rigorous mathematical and statistical treatment of methods for localizing genes and environmental effects involved in the etiology of complex traits using case-control and pedigree data. Prerequisites: BST 775; Knowledge of SAS and programming languages such as C++, and basic knowledge of multivariate methods and Markov chain theory is highly recommended. 3 hours. Spring (even years).

All DrPH students will complete an applied practice experience (Practicum) in which the student will complete at least one project that is meaningful for a public health organization and to advanced public health practice. In addition to self-reflection on the applied practice experience, the student will produce a final product that addresses the competencies listed below. This may take the form of a written report, portfolio, or other deliverable as determined by the student, advisor, and/or Practicum preceptor, according to departmental guidelines. Pass/No Pass. 3-6 hours.

This course provides an opportunity for post-doctoral students to learn about ongoing research in the field of biostatistics, clinical trials, and statistical genetics. Reserved for BST Postdoctoral students. Pass/No Pass. 3 hours. Fall/Spring

This course is designed to cover advanced special topics in Biostatistics that are not covered in regular 700 level courses, but suited for doctoral students in Biostatistics. Prerequisites: BST 622 and 632. Pass/No Pass. 1-3 hours.

Non-dissertation research with the guidance of appropriate faculty. Research conducted before admission to candidacy for the doctoral degree. Pass/No Pass. 1-12 hours.

Doctoral level dissertation research under the direction of the dissertation research committee. Prerequisite: Admission to candidacy for PhD Pass/No Pass. 1-12 hours.