CFRM 556 Statistical Modeling for Computational Finance

This course covers the theory and application of statistical models and machine learning methods commonly used in quantitative and computational finance. Students in the MS CFRM program take this course during their first quarter and are expected to be able to use these modeling methods in subsequent courses. Computing exercises are used to reinforce the theory and methods. Topics covered include the following:

  • Principal component analysis (PCA)
  • Least-squares linear regression model fitting and model selection
  • Modern shrinkage fitting methods for models with many predictor variables
  • Robust linear regression model fitting and robust PCA
  • Non-parametric regression methods
  • Applications to factor models for asset returns
  • Clustering and classification methods

Upon successful completion, students will be able to apply models from statistics and machine learning to a variety of financial applications, in particular to fixed income and portfolio optimization models.

 

Instructor: 
Kjell Konis
Textbooks: 
Hastie, Tibshirani and Friedman (2008). Elements of Statistical Learning, 2nd edition (download free PDF) and other textbook TBD
Software: 
R and R packages.
Prerequisites: 
CFRM 541 Investment Science, which may be taken concurrently, or equivalent or by permission.
Credits: 
4