Elective Course Descriptions

Electives (for minor requirements choose two electives from following)

Categorial Data Analysis

Applied Statistics 420: Categorical Data Analysis
This course represents an introduction to methods for analyzing categorical data. Methods covered include those for contingency-table data (i.e., all variables are nominal or ordinal), as well as regression models for nominal and ordinal outcome variables. Although distribution theory and maximum likelihood are introduced as needed, the emphasis is on learning when and how to apply the methods, and how to interpret the results. Computations will be done either by hand or with the SAS computer program. Previous experience with SAS is useful, but not assumed. PREREQ: ASTAT 350 or equivalent.  Course typically taught in Fall semester

Multilevel Modeling

Applied Statistics 430: Multilevel Modeling
Multilevel models (also called hierarchical, random-effects, and mixed-effects models) are an increasingly important statistical tool in many social sciences. Examples include education (data on students within schools), economics (panel data), political science (data characterized by states and years), law (police stops categorized by date, location, and ethnic group), medicine (meta-analysis), public health (small-area estimation), social work (studies of individuals within housing areas), and many other areas.

This course covers setup, inference, and checking the fit of multilevel models. Computation using the software packages R and Bugs and applications in social science and elsewhere. By the end of the course, you should be able to understand multilevel models and apply them creatively to your data-analysis problems. PREREQ: ASTAT 350 or equivalent.  Course typically taught in Fall semester

Factor Analysis and Related Methods

Applied Statistics 440: Factor Analysis and Related Methods
In factor analysis, a "factor" is an unobservable construct hypothesized to give rise to observed variables (e.g., responses to questionnaire items). This course introduces popular factor-analytic models and methods for fitting them to data, in both exploratory and confirmatory contexts. Models for (approximately) continuous observed data are covered, as well as those for categorical observed data, including a few models and methods of item response theory. Application and interpretation are emphasized, with statistical theory introduced as needed. Use of one or more computer programs will be required (prior experience with factor-analytic software is useful but not assumed). PREREQ: ASTAT 350 or equivalent.  Course typically taught in Spring semester

Panel Data

Applied Statistics 450: Panel Data
This course examines the significant statistical issues related to the analysis of panel data. Panel data can be generically described as containing multiple units observed at multiple points in time. Because panel data require attention to both heterogeneity and dynamics, we will cover both topics individually, in summary form, before considering their interaction and developing intuitions for situations that require greater attention to one than the other. Though a host of other topics will receive attention, we will focus on the following issues: (1) Can individual time series be pooled and under what conditions? (2) Deterministic vs. random sources of variation arising from units or time points; and (3) What issues arise in translating techniques for panel data to censoring, truncation, and other pathologies that result in limited dependent variables? Prerequisites: Intro to Applied Statistics (330/513), Intermediate Applied Statistics - Linear Models (350/515), the equivalent, or permission of the instructor.  Course typically taught in Fall semester

Time Series Modeling in Social Science

Applied Statistics 460: Time Series AnalysisThis course considers statistical techniques to evaluate social processes occurring through time.  The course introduces students to time series methods and to the applications of these methods.  Coverage will begin with the traditional ARIMA (Box-Jenkins) approach to time series analysis and proceed through dynamic modeling and regression approaches to recent developments such as cointegration analysis, error correction models, and vector autoregression.  We will learn not only how to construct these models but also how to use them in applied analysis.  Heavy emphasis will be given to fundamental concepts and applied work.  Prereqs for the course include a solid understanding of the fundamentals of statistical inference, regression analysis, matrix algebra, and the general linear model.  By the end of the course, you should be able to: (1) Use the Box-Jenkins modeling approach to prewhiten data and conduct an intervention analysis (2) Run and interpret time series models using econometric methods such as GLS and distributed lag models. (3) Analyze cointegrated data using an error correction model. (4) Use vector autoregression to analyze data and apply techniques such as impulse response and moving average response analysis to interpret results. PREREQ: ASTAT L55 364 or equivalent. Course typically taught in Fall semester.

Visiting Scholar Statistical Research Seminar in Applied Statistics

Applied Statistics 560: Visiting Scholar Statistical Research Seminar in Applied Statistics
This course brings distinguished academic statisticians to WU as part of an organized research seminar. Lacking a Statistics Department or Ph.D. program in statistics, the campus community can substantially benefit from internationally recognized scholars in the field who are willing to spend substantial time at the university. Selected statisticians will come to campus twice during the course. First, they will spend two days at the beginning of the semester to introduce a research topic in statistics and to assign a reading list of 8-12 technical papers, including some of their own authorship. Second, they will return to campus towards the end of semester for fur day for: two 2-hour seminars, a scholarly talk in the Center for Applied Statistics, and individual meeting time with seminar participants and other members of the university community. In-between these two visits, a faculty member in the Center for Applied Statistics will lecture and lead a discussion on each of these assigned papers as part of the weekly seminar meeting. The objective is to provide deep understanding of a complex technical topic through the use of experts in the field. PREREQUISITES: Math 439 or Political Science 581 or Biostat L24-439 or Econ 413, or approved equivalent.  Course typically taught in Fall and Spring semesters

Quantitative Political Methodology I

Applied Statistics 581: Quantitative Political Methodology I
Continuation of Psych 5066. Introduction to multiple regression/correlation analysis. Topics include bivariate and multiple correlation and regression, representation of nominal or qualitative variables, power and orthogonal polynomials, interactions, analysis of covariance, repeated measures design. PREREQ: Psych 5066.  Formerly Applied Statistics 515D: Intermediate Applied Statistics: Linear Models.  Course typically taught in Spring semester.  Same as home course Political Science 581: Quantitative Political Methodology I

Quantitative Methods II

Applied Statistics 5067: Quantitative Methods II
Continuation of Psych 5066. Introduction to multiple regression/correlation analysis. Topics include bivariate and multiple correlation and regression, representation of nominal or qualitative variables, power and orthogonal polynomials, interactions, analysis of covariance, repeated measures design. PREREQ: Psych 5066.  Same as home course Psychology 5067