STAT 406 2021W

Methods for Statistical Learning

Jump to Schedule.

Learning outcomes

At the end of the course, you will be able to:

  1. assess the prediction properties of the supervised learning methods covered in class;
  2. correctly use regularization to improve predictions from linear models, and also to identify important explanatory variables;
  3. explain the practical difference between predictions obtained with parametric and non-parametric methods, and decide in specific applications which approach should be used;
  4. select and construct appropriate ensembles to obtain improved predictions in different contexts;
  5. select sensible clustering methods and correctly interpret their output;
  6. correctly utilize and interpret principal components and other dimension reduction techniques;
  7. employ reasonable coding practices and understand basic R syntax and function.

Short list of covered topics

Flexible, data-adaptive methods for regression and classification models; regression smoothers; penalty methods; assessing accuracy of prediction; model selection; robustness; classification and regression trees; nearest-neighbour methods; ensembles; unsupervised learning.