FA631 Investment, Portfolio Construction, and Trading Analytics

Course Catalog Description


The significant amount of information available in any field requires a systematic and analytical approach to select the most important information and anticipate major events. Machine learning algorithms facilitate this process understanding, modeling and forecasting the behavior of major social or economic systems and their variables. This course explores how to apply fundamental machine learning models to predict financial time series and solve financial problems. Some of the financial applications explored are algorithmic trading, model calibration, portfolio optimization, and risk management.

Prerequisites: BIA656 Advanced machine learning and data analytics or FE590 Statistical learning in finance or FE690 Machine learning in finance or MIS 637 Data analytics and machine learning or CS559 Machine Learning or instructor’s authorization.

Campus Fall Spring Summer
On Campus
Web Campus


Professor Email Office
German Creamer

More Information

Course Outcomes

Students who successfully complete this course will be able to

Learning Goals:

  1. Apply statistical models and analytical methods to the finance domain.
  2. Recognize the value and limits of statistical learning algorithms to solve finance problems.
  3. Develop analytical models for financial forecasting, portfolio optimization and trading.

Course Resources


  • Trevor Hastie, Robert Tibshirani and Jerome Friedman, The Elements of Statistical Learning. Springer-Verlag, 2nd. Ed., New York, 2009
  • Marco López de Prado, Advances in Financial Machine Learning, Wiley, 2018 (LP)
  • Christopher D. Manning, Prabhakar Raghavan and Hinrich Schutze, Introduction to Information Retrieval, Cambridge University Press. 2008 (downloadable at http://nlp.stanford.edu/IR-book)
  • Selected papers:
  • G. Creamer and Y. Freund (2007). “A Boosting Approach for Automated Trading.” Journal of Trading 2 (3): 84-96.
  • G. Creamer (2015). “Can a Corporate Network and News Sentiment Improve Portfolio Optimization Using the Black Litterman Model?” Quantitative Finance 15 (8): 1405-1416.
  • G. Creamer (2012). “Model Calibration and Automated Trading Agent for Euro Futures.” Quantitative Finance 12 (4): 531-545.
  • S. Gu, B. Kelly, D. Xiu, Empirical Asset Pricing via MachineLearning, The Review of Financial Studies 33 (2020): 2223–2273.

Optional References

  • Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
  • Hal Daumé III, A Course in Machine Learning link


Grading Policies

HW 25% Participation 5% Research paper 35% Final exam 35%

Lecture Outline

Topic Reading Assignments
Week 1 Review of Python and data science concepts
Week 2 Classification models for risk management: decision trees, and support vector machine ESL 9.2 and 12 1: Python review
Week 3 Ensemble methods for algorithmic trading: bagging, boosting, and random forests ESL 7.1-7.3, 10.1-10.9, 15, 16 LP 6
Week 4 Technical analysis and algorithmic trading *1, *2 2: Classification: risk scores
Week 5 Clustering analysis, risk diversification and factor models ESL 14.3
Week 6 Natural language processing I: Sentiment analysis and financial trends MRS, 13-15
Week 7 Natural language processing II: Extracting information and investment signals from financial reports MRS, 13-15 3: Sentiment analysis
Week 8 Empirical asset pricing and machine learning I *3
Week 9 Empirical asset pricing and machine learning II *3 4: Text analysis and financial reporting
Week 10 Financial data structures and cross-validation LP 2-4, 7, ESL 7.10
Week 11 Feature importance & model calibration LP 8-9, *4
Week 12 Backtesting LP 11-14
Week 13 Bayesian models Mean variance portfolio optimization and Black Litterman model ESL 8.6, 17.1-17.3 *5
Week 14 Portfolio optimization LP 15-16 5: Portfolio optimization