STA 521

Predictive Modeling and Statistical Learning

An introduction to statistical methods for prediction, interpretation, and inference. This course introduces students to concepts and techniques of regression, including linear and generalized linear models, and topics in predictive modeling, including regularized methods, trees, kernel machines, and neural networks. The course will blend theory and application using a range of modern real-world examples and datasets. The R programming language and applications are used throughout.

Course info
Lecture M/W 1:25 - 2:40 PM Old Chemistry 116
Lab-01: Sylvia Tu 3:05 - 4:20 PM LSRC A247
Lab-02: Yiming Tu 4:40 - 5:55 PM Old Chemistry 201

Instructional team
Instructor   Yue Jiang T 2 - 3 PM; Th 11 AM - 12 PM Old Chemistry 208A
Teaching Assistant   Runxi Tang T 1:05 - 3:05 PM Old Chemistry 203A
Teaching Assistant   Sylvia Vincent T 10:00 AM - 12:00 PM Zoom ID 924 7811 7304
Teaching Assistant   Yiming Cheng M 3:00 - 5:00 PM Zoom ID 938 3113 6917
Header: We therefore hope it will be pleasing to geometers, if in this new treatment of the subject we have shown that the method of least squares presents the best possible combination. Not merely in an approximate sense, but in an absolute one, whatever may be the law of the probability of errors, whatever the number of observations, provided only that we establish the notion of the mean error not in the sense of the illustrious Laplace, but in the manner we have set forth in Articles 5 and 6. - Gauss, 1823.