Skip to main content

in this section

Financial Data Science @CFEM

FDS@CFEM is an immersion semester in Financial Data Science. It is currently offered as a fourth-semester option to our MFE students and results in the Certificate in Financial Data Science.  Students who are passionate about practical applications of large-scale data analytics, big data technologies, and predictive modeling will have a unique opportunity to deepen their knowledge in these subjects through our carefully–designed curriculum, created by world-renowned faculty and practitioners.  From mastering the theory of machine learning to implementing big data technologies to solving real large-scale finance problems, FDS@CFEM will move students beyond traditional quantitative finance into the realm of true value creation through the use of analytics. Companies, from global banks, to hedge funds, to fintech start ups, are increasingly focusing on the best way to take advantage of machine learning applications.  Whether you are interested in trading or risk management, asset management or modeling, regulation or policy-making, FDS@CFEM gives you invaluable tools to become the financial engineer of the future.   

FDS@CFEM students take courses to deepen their technical knowledge in machine learning algorithms and big data technologies. These foundational courses are integrated with hands-on project work where students learn directly from practitioners working in the field of financial data science. 

To learn more about the CFEM student experience, click here.

FDS@CFEM Spring 2018 Coursework

ORIE 5260: Machine Learning in Finance; 4 credits

Instructor: Neal Parikh, a Visiting Lecturer at Cornell University and Co-Founder of SevenFifty, an NYC-based technology company providing a data and web platform to the wholesale beverage alcohol industry.

Machine learning is a field at the intersection of computer science and statistics that aims to develop computational systems that learn from data and improve with experience. Though its origins lie in the field of artificial intelligence, modern machine learning has transformed a huge variety of areas, such as biology, medicine, retail, marketing, operations, logistics, politics, journalism, and, of course, finance.

This course provides a general introduction to machine learning with a view towards applications in finance. The goal is to provide both a solid grounding in the mathematical foundations of machine learning as well as a conceptual map of the field and its relation to areas like statistics and optimization that are currently more familiar in finance. The emphasis is on mathematical understanding, not implementation or financial specifics.

Sample topics include generalized linear models, loss functions and regularization, sparsity, support vector machines, kernelization, principal components analysis, clustering, and the

EM algorithm. Distinctions between classes of methods, such as probabilistic vs. variational models, Bayesian vs. frequentist approaches, and convex vs. nonconvex models.

There will be problem sets throughout the semester.

New for Spring 2018!

ORIE 5265 Financial Data Science Seminar; 2 credits

Instructor: Jeff Yau, Chief Data Scientist at Alliance Bernstein

With the massive growth of data and rapid development in computational technologies, the finance industry has experienced a paradigm shift towards the use of quantitative methods in all areas of finance. This shift makes machine learning an essential tool for modern quantitative finance professionals. This course applies some of the most popular machine learning techniques to different problems frequently encountered in the investment management industry and the banking industry. The applications will include, but not limited to, stock selection, investment strategy construction, credit risk forecasting, and sentiment analysis on corporate filings or central bank transcripts.  This course emphasizes the vigorous application of machine learning methods; this means that students will have to master the theoretical underpinning of the techniques before correctly applying them. As such, students taking this course are required to take ORIE 5260 (Machine Learning for Finance) as a co-requisite. Successful completion of this and ORIE 5260 will equip students with one of the most important quantitative tools for modern quantitative finance professionals. The course will consist of a series of evening lectures and students will have to complete 3-4 min project labs.

ORIE 5270: Big Data Technologies; 2 credits

Instructor: Saul Toscano Palmerin, ORIE

This course offers a broad overview of computational techniques and mathematical skills that are useful for data scientists. The course is designed as a “boot camp” to offer students an intensive immersion into the subject over a short period of time. 

The material will be taught during 1/16/18 - 1/27/18, 9am-4pm with 12-1pm lunch break; no class on Sunday. There will be recitation-style assignments each day, and on-going projects throughout the course. All work will be submitted by creating a web service on the course server.

The topics include:

  • Unix shell
  • Version control: Git
  • iPython
  • Creating web APIs
  • Data Structures and Algorithms 
  • Working with databases
  • Exploratory data analysis: using Python and related libraries to explore data sets (pandas, bokeh)
  • Map-Reduce, Spark, Hadoop.
  • Overview of some machine learning and optimization algorithms (logit regression, Poisson regression, k-means, neural networks, stochastic gradient descent, gradient descent, lbfgs)
  • Python libraries for data analysis (scikit-learn, pytorch, SciPy, numpy)
  • Parallel computing
  • Unit testing
  • IEEE 754 (Infinity, NaN, rounding error, overflow and underflow)

ORIE 5280 Big Data Practicum; 3 credits

Instructors: Dr. Sasha Stoikov, CFEM, Marcos Lopez de Prado, Guggenheim, Tarun Gupta and Arseniy Kukanov, AQR

The course covers practical applications of data science to finance. Three financial practitioners bring forth large data sets and problems where machine learning methods can be applied. Students will complete homework and a project.

The topics include:

  • Financial data management (sampling, labeling, weighting, fractional differentiation)
  • Modelling (ensemble methods, cross-validating (CV) in finance, feature importance, hyper-parameter tuning
  • Backtesting (bet sizing, historical backtests, synthetic data, strategy risk)
  • ML-based asset allocation
  • Financial features (structural breaks, entropy features, microstructural features)
  • How to design, develop and implement quantitative investment strategies. Alpha models: momentum, value and quality factors. Backtesting.
  • Risk models and mean-variance optimization
  • Transaction costs modeling
  • Market impact estimation
  • Trade execution