STATS 3DA3 - Data Science Methods

3 unit(s)

This course covers a range of statistical and machine learning methods, including classification, clustering, decision trees, random forest, bagging and gradient boosting for trees and linear models, ridge regression, LASSO, generalized additive models, principal component analysis (singular value decomposition), multiple hypothesis testing, sensitivity and specificity analysis, cross-validation, and bootstrapping. Python will be the primary software used, with R and other environments also used at the discretion of the instructor. This course includes a scientific communication component.
Two lectures, one lab; one term
Prerequisite(s): STATS 2DA3; one of MATH 1MP3, COMPSCI 1MD3, or PNB 2A03; one of ARTSSCI 2R03, STATS 2B03, or STATS 2MB3; and credit or registration in one of MATH 2LA3 or MATH 2R03