Multivariate Analysis with Machine Learning

This course is on multivariate statistical analysis with machine learning. Topics include an introduction to the mathematics to understand multivariate analysis and machine learning, most notably linear algebra, along with methods such as principal components analysis, correspondence analysis, multidimensional scaling, clustering, and discriminant analysis, as well as considering inference methods such as jackknifing, permutation, bootstrapping, cross-validation, and regularization. Insofar as many machine learning methods build on these techniques, a larger goal of the course is to prepare students for using larger scale extensions that exist in machine learning by incorporating modern extensions of classical methods as well as using contemporary examples drawn from text analysis, genomics, biometrics, neuroscience, and social science data analysis. Statistical computing and visualization methods using R are incorporated throughout. EPSY 70600 (Statistics II) or equivalent is a prerequisite; EPSY 74000 (Mathematical Foundations for Social Scientists) or other experience with mathematics is recommended.

Offer as needed

030893