Engineering Management Courses
Introduction to Computational Finance – Fall
Introduces the metrics and models used in the finance industry to quantify the potential performance and risk of investments, to construct investment portfolios that optimize risk-adjusted returns, and to evaluate the performance of money managers, in both traditional and alternative asset classes.
Students also learn the vocabulary, and experience the wins and losses of trading, through use of the Rotman trading simulation software. The Rotman simulation supports a number of realistic case studies, including market-making for equities, identifying arbitrage opportunities in commodities, and algorithmic trading. Students can also prepare for the Rotman Trading Competition Team Tryouts - see below.
The class assumes no prior knowledge of finance and all models and problem sets are completed in MS Excel. The course prepares students for Advanced Financial Engineering, taught in the Spring.
Advanced Financial Engineering – Spring
A seminar for students who are familiar with computational finance methods and concepts. The focus is on analysis of real markets, and the identification, optimization, and back-testing of algorithmic trading strategies. No prior programming skills required.
Data Mining – Fall
Intended to as an introduction to the fundamental building blocks of practical data-analysis: business metrics, models and algorithms, and measures to evaluate and rank model effectiveness. Given raw data, students learn to ask questions, do exploratory data analysis, develop models, and emerge with persuasive recommendations for business process change. The course has a strong information theory orientation – we approach data mining as the process of identifying and amplifying an actionable signal in an ocean of noise. At least half the Semester is devoted to individual and team project work on real-world data sets.
New Opportunities in Big Data
New technologies and algorithms are transforming the fields of machine learning and predictive analytics. Real-time customization of user experiences, for example, requires careful planning regarding what data to pre-process, what algorithms to run in real time, and how to integrate new information into prior models for maximum information gain. Insight into these topics is sought using the Bayesian Logical Data Analysis approach, a rigorous perspective on information theory and algorithmic learning pioneered by E.T. Jayes, David MacKay, D.S. Sivia, Phil Gregory, and others.
This course has a highly practical side also - students typically work throughout the semester as consultants to high-tech companies that need answers to data-analytic questions.
Students also prepare for, participate in, and on occasion win, the national DataFest Competition held each March. In DataFest, which companies make large commercial data sets available for business analysis over a weekend. Past data-sponsors include Edmunds, GridPoint, eHarmony, and Kiva.com. DataFest is organized at Duke by Dr. Mine Çetinkaya-Rundel from the Duke Statistics Department.
Rotman Trading Competition
Each February, the Rotman School of Business in Toronto Canada hosts an International Trading Competition. This event brings together students from the top Business Schools around the world. Duke has fielded a team every year since 2011. Tryouts for the Duke team, which has six members, are typically held in October, after scheduled practice trading events. The team is open to all currently-enrolled Duke students. Taking the Computational Finance course is not required.
Students practice a critical business skill: analyzing real-world data and communicating actionable findings in compelling form. This is a small group, project-based class. Students can design their own project or contribute to a CQM project that is already underway.
In Fall 2015, Egger and collaborator Jana Schaich-Borg are launching a new Coursera four-course Specialization in data science, called Excel to MySQL: Business Analytics Techniques
The individual courses and their launch dates are as follows.
- Course 1: Business Metrics for Data-Driven Companies – September 15, 2015
- Course 2: Mastering Data-Analysis in Excel – October 15, 2015
- Course 3: Data Visualization and Effective Communication with Tableau – November 15, 2015
- Course 4: Mastering Big Data with MySQL – December 15, 2015
Verified Certificates are available from Coursera in each course as well as for a Capstone Project that will give students an opportunity to demonstrate effective working knowledge of the material from all four courses.
Please click here to access the courses.
Duke Students who wish to pursue their own independent research in Data Analytics may be eligible for course credit through a supervised Independent Study project. If interested, please contact Daniel Egger directly.