Statistics (STAT)

Description: Statistical concepts and statistical methodology useful in descriptive, experimental, and analytical study of biological and other natural phenomena. Practical application of statistics rather than on statistical theory.

Description: Statistical concepts and methodology useful for description, analysis and interpretation of experimental and observational studies. Practical application of statistics and essential background for subsequent courses in statistics.

Description: Essential statistical characteristics of a research study intended to assess the impact of treatment, environmental or population conditions on response. Focus is on both designed experiments and on studies for which controlled experiments are not feasible but characteristics of controlled experiment must be mimicked to the extent possible. Methods to assess power and compare efficiency of alternative designs are considered. Course covers major design structures, including blocking, nesting, multilevel models, split-plot and repeated measures, and statistical analysis associated with these structures.

Description: Model-based inference for ecological data, generalized linear and additive models, mixed models, survival analysis, multi-model inference and information theoretic model selection, and study design.

Description: Sampling techniques: simple random sampling, sampling proportions, estimation of sample size, stratified random sampling, ratio and regression estimates.

Description: Program requirements, resources available, tips for academic success, professional statistical organizations, career paths, history of statistics, ethics, statistical conferences, statistical blogs and online forums, frequentist and Bayesian paradigms, current research in department.

Description: Designed primarily to develop and equip middle-level teachers with the statistical knowledge they need for teaching. The course follows an inquiry/discovery design, dedicating much of class time to activities, discussion and group work. The course emphasizes topics in statistics that are part of the middle-school mathematics curriculum, as well as their application in other disciplines. The course also includes statistics that are used in education and school-based research.

Description: The statistical concepts typically taught in a high school statistics class, including linear regression, two-way tables, sampling distributions, statistical inference for means and proportions, chi-square tests, and inference for regression. Some experience with basic statistical concepts (mean, standard deviation, elementary probability) is necessary. The course is inquiry-based, and will emphasize applications and statistical thinking.

Description: Introduction to essential statistical methods and supporting design and modeling theory for professional statistical practice. First in a three semester sequence. Focus of this course on methods for single response variable and non-hierarchical study design.

Description: A continuation of Statistical Methods I. Second in a three semester sequence on essential statistical methods and supporting design and modeling theory for professional statistical practice. Focus in this course of methods for single response variable and multiple sources of random variation.

Description: Introduction to essential statistical methods and supporting design and modeling theory for professional statistical practice. Third in a three semester sequence. Focus of this course on methods for situations that extend beyond the single-response-variable, designed study cases featured in Statistical Methods I and II. These include multivariate statistics, non-linear models, non- and semi-parametric statistics, observational studies, and other theory and methods deemed appropriate as statistical science continues to evolve.

Description: Introduction to the role and purpose of statistical consulting and interdisciplinary collaboration. Topics include: asking good questions, dealing with difficult clients, communicating statistics to non-statisticians, determining solutions, and collaborating.

Description: Food evaluation using sensory techniques and statistical analysis.

Description: Statistical methods for modeling and analyzing correlated data, with emphasis on spatial correlation. Descriptive statistics, time series, correlograms, semivariograms, kriging and designing experiments in the presence of spatial correlation.

Description: Statistical methods useful for analyzing sports-related data. Descriptive statistics, graphical representations, experimental design, discriminant analysis and optimization.

Description: Basic biological concepts. Multiple testing and false discovery rate. Second generation sequencing and statistical issues. ChIP-seq. RNA-seq. Empirical Bayes methods and software. Normalization, experimental design and commonly used models for microarray data. Metabolomics.

Description: Databases, high-throughput biology, literature mining, gene expression, next-generation sequencing, proteomics, metabolomics, system biology and biological networks.

Description: Next-generation RNA and genome sequencing, systems biology. Regulatory networks of transcription, protein-protein interaction networks, theory and practice. Databases, data integration and visualization. Students present computational biology publications and projects.

Description: Quantitative genetic analysis of complex traits. Quantitative methodologies for connecting phenotypes with high-dimensional genomic information to understand polygenic traits from both prediction and inference perspectives.

Description: Theoretical concepts involved in planning breeding programs for the improvement of measurable morphological, physiological, and biochemical traits that are under polygenic control in crop plants of various types.

Description: Introductions to statistical computing packages and document preparation software. Topics include: graphical techniques, data management, Monte Carlo simulation, dynamic document preparation, presentation software.

Description: Principles in the estimation of (co)variance components and genetic parameters required to solve mixed models typical in livestock genetics. Focus on applied knowledge of approaches used to estimate the G and R sub-matrices of the mixed model equations. Demonstrate models commonly used in parameter estimation. Introduce scientific literature concerning implementation, and attributes of the solutions, of variance component estimation strategies.

Description: Introduction to the R environment for statistical computing, including use of R as a high-level programming language and as a gateway for more formal low-level languages. Material includes language structure, basic and advanced data manipulation, statistical analysis with R, and using R as a programming language.

Description: Principles of Markov Chain Monte Carlo (MCMC) methods in animal breeding. Materials include random variable generation, Monte Carlo integration, stochastic search, Expectation-maximization (EM) algorithm and Monte Carlo EM, Markov Chain principles, Metropolis-Hastings algorithm, Gibbs sample, and MCMC for genomic data. Illustrations developed using R software.

Description: Linear regression and related analysis of variance and covariance methods for models with two or more independent variables. Techniques for selecting and fitting models, interpreting parameter estimates, and checking for consistency with underlying assumptions. Partial and multiple correlation, dummy variables, covariance models, stepwise procedures, response surfaces estimation, and evaluation of residuals.

Description: Methods and underlying theory for analyzing data based on generalized, linear, and mixed models.

Description: Multivariate techniques used in research. Reduction of dimensionality and multivariate dependencies, principal components, factor analysis, canonical correlation, discriminant analysis, cluster analysis, multivariate extensions to the analysis of variance, and the general linear model.

Description: Introduction to nonparametric statistics - methodology and supporting theory. Focus of this course is broadly divided into three components: traditional (e.g. distribution-free hypothesis testing), function estimation (e.g. alternatives to parametric linear and nonlinear models) and modern methods that emphasize prediction (e.g., density estimation, robustness, computational methods, reproducing kernel Hilbert space methods).

Description: Analysis of contingency tables. Regression models for binary, multi-category, and count responses. Tools for model building. Exact inference methods.

Description: Application, theory and computational aspects of survival analysis. Survival and hazard functions; parametric models for survival data; censoring and truncation mechanisms; nonparametric estimation (confidence bands for the survival function, interval estimation of the mean and median survival time); univariate estimation of the hazard function; hypothesis testing; regression models (with fixed covariates, with time dependent covariates); and model diagnostics.

Description: Practical application of mixed models for data analysis, estimation, prediction, and testing. This course covers linear mixed models (LMM) for normally distributed data and generalized linear mixed models (GLMMs) for non-normally distributed data.

Description: Introduction to models for data observed over time. Both theoretical and practical aspects of time series models will be presented. Main topics include the Box-Jenkins model class, spectral analysis, and GARCH models. Forecasting will be emphasized throughout. The main statistical software package used will be R. Other statistical software packages and programming languages will be introduced as needed.

Description: Introductory mathematical statistics. Probability calculus; random variables, their probability distributions and expected values; sampling distributions; point estimation, confidence intervals and hypothesis testing theory and applications.

Description: Sample space, random variable, expectation, conditional probability and independence, moment generating functions, special distributions, sampling distributions, order statistics, limiting distributions and central limit theorem.

Description: Interval estimation; point estimation, sufficiency and completeness; Bayesian procedures; uniformly most powerful tests, likelihood ratio test, goodness of fit tests.

Description: Introduction to stochastic modeling in operations research. Includes the exponential distribution and the Poisson process, discrete-time and continuous-time Markov chains, renewal processes, queueing models, stochastic inventory models, stochastic models in reliability theory.

Description: The key topics represent the main areas of data mining and machine learning aimed at achieving predictive accuracy more than physical modeling. These topics are standard classification methods, regularization methods, visualization and geometry of data, leading to kernel methods. Finally, the course introduces trees, neural nets, and model averaging.

Description: Bayesian data analysis with emphasis on application and computation using R or similar software. Topics include: probability models, prior distributions, Bayes theorem, single parameter models, posterior predictive distribution, Gibbs sampling, MCMC simulations, regression models, generalized linear models, hierarchical models, model checking and diagnostics.

Description: Special topics in either statistics or the theory of probability.

Description: A student may take this class and prepare a Statistics report under the supervision of his/her faculty advisor.

Description: Advanced design concepts, theory and methods used in: construction, analysis and interpretation of incomplete block designs, split-plots, confounded and fractional factorials, screening designs, response surface methods, and other topics.

Description: Exposure to more complex statistical consulting problems and how to resolve them. Topics include: major areas of consulting, interdisciplinary collaboration, and effective communication. Students will assemble a portfolio of project reports that can be shared with prospective employers.

Description: This course provides a development of theory and methods for spatial and spatio-temporal statistics. It provides the mathematical foundations and methodological development for topics such as MLE and in-fill asymptotics, non-Gaussian/non-stationary spatial processes, spatio-temporal models, and Bayesian methodology. R will be the main programing language.

Description: Statistical computing needed for research and advanced statistical analyses. Topics include: bootstrap, high performance computing, jackknife, Linux, Markov chain Monte Carlo, Monte Carlo simulation, numerical differentiation and integration, optimization, parallel processing, permutation tests.

Description: A continuation of Computational Statistics I. Topics will be chosen from big data management and data analysis, data generation, high performance and throughput computing, importance sampling, machine learning, optimization, programming languages, web scraping, working with databases.

Description: Statistical inference concerning parameters of multivariate normal distributions with applications to multivariate datasets.

Description: Basic concepts of nonlinear models and their associated applications. Estimating the parameters of these models under the classical assumptions as well as under relaxed assumptions. Major theoretical results and implementation using standard statistical software.

Description: Construction of probability spaces, random variables and expectations, monotone and dominated convergence theorems, Fatou's lemma, modes of convergence, Kolmogorov law of large numbers, central limit theory, conditional probability given a sigma field.

Description: A continuation of STAT 980 providing depth in probability theory and stochastic processes. Topics include convergence properties of random variables and treatment of several important stochastic processes.

Description: Uniformly minimum variance unbiased estimators, decision-theoretic Bayes estimation, frequentist testing (likelihood ratio tests, Neyman-Pearson lemma, uniformly most powerful tests), Bayes testing and Bayes factors, nonparametric tests, multiple comparisons procedures.

Description: Model selection including sparsity methods and their oracle properties, information methods, cross-validation and stochastic search. Basic theory of kernel methods for regression. Classification: linear and quadratic discriminants, Bayes classifier, nearest neighbor methods, kernel methods for classification. Introduction to neural networks and recursive partitioning. Model averaging methods and measures of complexity. Cluster analysis.

Description: A continuation of STAT 980 providing breadth in commonly occurring major subfields of statistics that rely heavily on probability theory. Large sample theory estimation, testing, expansion, and convergence in a variety of settings.

Description: The foundational ideas and structure of Bayesian theory from its axiomatic and fundamental assumptions, including Savages axioms, complete class theorems, sequential properties, prior selection, model selection, Bayesian nonparametrics, and asymptotics for both the parametric and nonparametric cases.

Description: Special topics in either statistics or probability.

Description: Participation in the activities of a practicing statistician.

Description: Doctoral Dissertation Research