Fall 2015/Sp 2018: BIOS 625 [Categorical Data Analysis & Generalized Linear Models]
Semester course; 4 lecture hours. 4 credits. Topics include exact and asymptotic analysis of contingency tables; measures of association and agreement; theory and applications of generalized linear models, maximum likelihood estimation and related numerical methods; linear models with different link functions and distributions; model fitting; and diagnostics. The course includes hands-on experience using software, such as SAS and R.
Fall 2016: BIOS 667 [Statistical Learning & Data Mining]
Semester course; 3 lecture hours. 3 credits. Specific topics will include discrimination analysis, k-nearest neighbors, naive Bayes classifiers, classification and regression trees, ensemble methods, random forests, L1 penalized models, bootstrap and cross-validation methods. The course includes hands-on experience using statistical software for each method, mostly in R.
Sp 2017: BIOS 546 [Theory of Linear Models]
Semester course; 3 lecture hours. 3 credits. Specific topics will include review of linear algebraic concepts and matrix operations, generalized inverses and systems of equations, distribution of quadratic forms under normal theory; general linear models of full rank and less than full rank; least squares and maximum likelihood estimation; hypothesis testing; multiple linear regression; analysis of variance; balanced and unbalanced designs. The course includes hands-on experience using the R software.
Sp 2019/Fall 2020/Sp 2022/Sp 2023/Sp 2025: BIOS 647 [Survival Analysis]
Semester course; 3 lecture hours. 3 credits. Specific topics will include the analysis of survival (or failure time) data, with/without censoring. actuarial and life-table methods, nonparametric and parametric estimation of survival functions, and comparison of survival curves; regression methods, such as the Cox proportional hazards model and accelerated failure time models; competing risks; interval censoring and current status data; multistate models; joint modeling of longitudinal and survival data. The course includes hands-on experience using the R software.
Sp 2020: BIOS 691 [Special Topics in Biostatistics: Advanced Bayesian Analysis]
Semester course; 3 lecture hours. 3 credits. Specific topics will include Bayes decision theory; interpreting posteriors; point and interval estimation; hypothesis testing; Bayesian hierarchical modeling; Empirical Bayes methods; Asymptotic methods and approximations; Bayesian computing, with emphasis on big data; Bayesian approaches for experimental designs and clinical trials; meta-analysis and model averaging; mixture models; Bayes nonparametrics; Bayesian case studies. The course includes hands-on experience using the R software.
Fall 2021/2022/2023: BIOS 601 [Analysis of Biomedical Data]
Semester course; 3 lecture hours. 3 credits. This course provides an overview of the analysis of continuous response data. The material begins with a brief review of theoretical tools used in inference and segues into common univariate and bivariate statistical methodologies for the analysis of continuous response data. Model-based statistical methods including linear regression, ANOVA, ANCOVA and mixed-effect models will also be covered. Practical consideration and usage of statistical methods, utilizing commonly used statistical software packages, will be emphasized over theoretical underpinnings of the methods. The course includes hands-on experience using software, such as SAS and R.
Fall 2024: BIOS 549 [Spatial Data Analysis]
Semester course; 3 lecture hours. 3 credits. This course is designed as an introductory spatial course, and covers spatial data visualization and manipulation, spatial point pattern analysis, interpolation and geostatistics for point-referenced data, and spatial regression modeling of areal data. Includes the use of a statistical software package for data analysis. The course includes hands-on experience using the R software.
Fall & Sp 2016/2017/2018/2019/2020/2021/2022/2023/2024: BIOS 690 [Biostatistical Research Seminar]
Semester course; 1 lecture hour. 1 credit. Seminars (either in-person, or zoom) will be presented mostly by external speakers on Fridays, approximately twice a month, and students are required to submit their understandings of 3 presentations of their choice as 3 homeworks.
--------------------------------------------------------------------------------------------------------------------------------------------------------------
Sp 2012/2014/2015: PubH 8472 [Spatial Biostatistics]
Semester course; 3 lecture hours. 3 credits. This course is designed as an advanced spatial data course, with an emphasis towards Bayesian treatments, and covers fundamentals of catrography, point-referenced data analysis (variograms, details on stationarity and isotropy, classicial estimation techniques and theory, kriging), areal data (markov random fields, conditionally/simultaneously autoregressive models), hierarchical modeling for univariate spatial data, Bayesian kriging, spatial misalignment, point and block level modeling, multivariate spatial modeling, separable point level models, co-regionalization models, spatio-temporal modeling (with/without alignment), spatial survival modeling (frailty and cure rates), spatially-varying co-efficients, spatial CDFs, wombling, spatial big data. Includes the use of R statistical software packages for data analysis. The course includes hands-on experience using the R software.
Sp 2013/2014/2015: PubH 7440 [Introduction to Bayesian Analysis]
Semester course; 3 lecture hours. 3 credits. This course is designed as an intermediate Bayesian course, and covers theory and application of Bayesian linear models, details on Bayesian computing (Gibs sampling, Metropolis-Hastings algorithm, slice sampling), Bayesian model choice and assessments, empirical Bayes methods, frequentist comparisons, Bayesian design and analysis of clinical trials, hierarchical longitudinal models, Bayesian survival modeling with frailty, spatio-temporal modeling, and some Bayesian case studies. Includes the use of software, such as R, WinBUGS, OpenBUGS and JAGS.
--------------------------------------------------------------------------------------------------------------------------------------------------------------
Sp 2007/2011: BMTRY 711 [Analysis of Categorical Data]
Semester course; 3 lecture hours. 3 credits. Topics include exact and asymptotic analysis of contingency tables; measures of association and agreement; theory and applications of generalized linear models, maximum likelihood estimation and related numerical methods; linear models with different link functions and distributions; model fitting; and diagnostics. The course includes hands-on experience using software, such as SAS and R.
Sp 2008: BMTRY 718 [Stochastic Processes in Biology and Medicine]
Semester course; 3 lecture hours. 3 credits. Topics include some initial lectures on probability theory, followed by theory and biological applications of discrete time Markov chain (such as, gambler's ruin, birth and death processes and epidemic processes), branching process, theory and biological applications of continuous time Markov chain, birth and death processes, applications to epidemic, and population genetics, continuous time and state Markov processes (such as diffusion process), and stochastic sifferential equations. The course includes some hands-on experience using the R software.
Fall 2009: BMTRY 704 [Nonparametric Methods in Biology and Medicine]
Semester course; 3 lecture hours. 3 credits. Topics include understanding fundamental differences between parametric and nonparametric methods (in terms of power and efficiency, appropriateness, etc), one sample cases (Binomial test, Kolmogorov-Smirnov tests, run test), two-related samples (McNemar's test, sign test, Wilcoxon test and Walsh test), two independent samples (Fisher's exact probability test, Chi-square test for independent samples, median test, Mann-Whitney test, Kolmogorov-Smirnov test and Wald-Wolfowitz test), k-samples (Cochrane's Q test, Friedman's two-way ANOVA), k independent samples (Chi -square test, Kruskal-Wallis one way ANOVA), and nonparametric correlation (contigency coefficient C, Spearman's rank correlation, Kendall's rank correlation, Kendall's partial correlation coefficient). The course includes some hands-on experience using the R software.