We will not have all the answers to many of the questions we address in the course. Our goal is teach students how to think about the issues, rather than what to think. The principles we aim to impart on students, taught through the lens of sports, will be useful in addressing many business-related issues, even outside of sports.
This course is designed to help students develop and apply analytical skills that are useful in a general business environment, applied specifically to sports. Students will learn how to apply methods and principles in a wide range of applications: evaluating performance and decision making, hypotheses testing, interpreting market-based evidence, identifying directions of causation, and quantifying the magnitude of various effects. Although the course focuses on applications in the sports context and uses approaches that are rapidly becoming important in the business of sports analytics, this is not a survey course about issues in sports.
Probability modeling and statistics are fundamental tools of management. Building upon earlier core courses in probability, statistics, economics, and modeling managerial decisions, Sports Analytics is designed to provide additional experience using these modeling tools via applications to problems in sports. Specific analytical tools include conditional probability, conditional expectation, random variables and probability distributions, hypothesis testing, and regression (including identification strategies such as instrumental variables, as well as logistic regression). We will consider various applications to major sports such as baseball, basketball, football, hockey and soccer. Sample applications include in-game decision making, rating sports teams and individual players, modeling win probability and win probability added in various sports, understanding the determinants of home field advantage, understanding referee behavior, sports gambling, and player evaluation. Students will gain practice applying analytical tools to these topics via regular problem sets and a term project.
Applied Quantitative Finance
This course develops, examines, and applies models for portfolio decisions by investors and the pricing of securities in capital markets. While developing portfolio theory, we will study the extensive empirical work that characterizes movements in security prices, evaluates alternative investment and asset pricing models, and attempts to test those models and interpret the implications of those tests. This is a research-oriented course with practical implementation of quantitative methods in finance, aimed at highly motivated and technically proficient MBA and undergraduate students.
This course is designed for students who want a more detailed and up-to-date treatment of academic asset pricing theory and empirical work and its application to the practice of quantitative finance. The course is especially appropriate for students contemplating analytical finance and quantitative money management and provides many tools and concepts that are essential for a career in quantitative investments. The material is covered in a rigorous analytical manner, and students must be comfortable with some technical methodologies (i.e., calculus, linear and matrix algebra, and statistical theory). The course is meant to be
challenging, but accessible. The expectation is that the average student spends 10-20 hours per week on the course, outside of class.
A good fundamental background in economics and especially statistics is required. The course is highly quantitative because the field is, and so relies heavily on analytical tools and economic theory developed throughout the course. Students should be comfortable with probability, statistics, and regression analysis. Students should also feel comfortable with the concepts of risk aversion, utility functions, and budget constraints. Use of a statistical package or programming language will be vital for the course, saving time and aiding in understanding the material. Many of the applications will move beyond simple spreadsheet packages such as Excel. A good statistical programming language such as Matlab, SAS, Stata, Python is even more useful. I will supplement the course with programming help and the data assignments will be done in groups.