This course is designed for students who wish to increase their capability to build, use, and interpret statistical models for business. It builds on the statistical background gained from B6014, the core course in managerial statistics. Students with questions about the course are encouraged to contact the professor at [email protected].

A primary goal of the course is to enable students to build and evaluate statistical models for managerial use in finance, operations and marketing. The focus is on generating managerially useful information and practical decision-making tools, rather than on statistical theory per se. A number of actual business cases are studied.

Concepts covered are multiple linear regression models and the computer-assisted methods for building them, including stepwise regression and all subsets regression. Emphasis is placed on diagnostic and graphical methods for testing the validity and reliability of regression models.

Course topics include a review of basic statistical ideas, numerical and graphical methods for summarizing data, simple linear and nonlinear regression, multiple regression, qualitative independent and dependent variables, diagnostic methods for assessing the validity of statistical models. The course studies applications of regression to business forecasting and also examines alternative times series forecasting models, including exponential smoothing.

While the primary focus of the course is on regression models, some other statistical models will be studied as well, including cluster analysis, discriminant analysis, analysis of variance, and goodness-of-fit tests.

Term project: A major aspect of course is the opportunity to carry out a practical statistical analysis project of one’s own. Students work in teams on a problem of their own choosing. The goal of the project is to develop a useful statistical model for a specific business problem, with the professor providing ongoing guidance and advice during the course of project. The teams will give an oral presentation of their results at the term’s end. Examples of previous student projects may be found at http://www.columbia.edu/~dj114/8899projects.htm.

Excel is used for basic statistical analysis as well as for developing straightforward regression models. In addition, more advanced commercial statistical software, such as Minitab or SAS, is used to carry out more complex and advanced analyses. In addition to the term project, there will be several computer-based assignments.

Paul Glasserman

Jack R. Anderson Professor of Business

Professor Glasserman's research and teaching address risk management, derivative securities, Monte Carlo simulation, statistics and operations. Prior to joining Columbia, Glasserman was with Bell Laboratories; he has also held visiting positions at Princeton University, NYU, and the Federal Reserve Bank of New York. In 2011-2012, he was on leave from Columbia and working at the Office of Financial Research in the U.S...

View full profilePersonal Website