MGTSC - Management Science

An overview of multivariate data analysis normally taken by students in the first year of the Business PhD program. Designed to bring students to the point where they are comfortable with commonly used data analysis techniques available in most statistical software packages. Students are expected to complete exercises in data analysis and in solving proofs of the major results. Topics will include univariate analysis, bivariate analysis, multiple linear regression, and analysis of variance. It is expected that students have as background at least one semester of calculus, one semester of linear algebra, and two semesters introduction to probability, probability distributions and statistical inference. Prerequisite: Registration in Business PhD Program or written permission of instructor. Approval of the Business PhD Program Director is also required for non-PhD students.

A continuation of the overview of multivariate data analysis begun in MGTSC 705. Topics include categorical data analysis, multivariate linear regression, discriminant analysis, canonical correlation, multivariate analysis of variance, principal component analysis, factor analysis, cluster analysis and logistic regression. Prerequisite: MGTSC 705 or consent of Instructor. Approval of the Business PhD Program Director is also required for non-PhD students.

This course is organized into two parts. Part I covers univariate and multivariate time domain models of stationary and nonstationary time series. Topics covered include univariate time series models, unit root tests, time series regression modeling, systems of regression equations, vector autoregressive models for multivariate time series and cointegration. In Part II the course introduces the issues and opportunities that arise with panel data and the main statistical techniques used for its analysis. Topics covered include fixed-effects models, random-effects models, dynamic models and limited dependent variable models. Throughout the course, the emphasis will be on how to use S-plus and Stata to estimate panel data and time series models. There is relatively less emphasis on statistical theory. Evaluation in the course is based on home work assignments and a term project. Prerequisite: MGTSC 705 or equivalent.