Regression Graphics for Advanced Statistical Modeling

3 Units

This course covers model-based and model-free techniques for extracting regression information in generalized linear models, single index models, and beyond. The course will introduce unsupervised dimension reduction including principal component analysis (PCA) and other non-linear techniques and their use in a supervised setting known as sufficient dimension reduction (SDR), for visualizing linear and nonlinear regression relationships. The techniques covered in this course are applicable in dealing with high dimensionality problems in machine learning and artificial intelligence. Throughout the course, theorems, algorithms, and practical applications will be covered to guide students in the process of analyzing regressions graphically, assessing, and selecting models. Prereq: STAT 325 or STAT 425 or STAT 445 .