Statistical Learning in Data Science

Catalog description

This course provides detailed overviews over the evaluation and application of statistical learning theories and techniques for inference and prediction in data science, particularly for biological and public health data. Topics include linear and nonlinear models, resampling techniques, tree-based methods, unsupervised learning such as clustering, support vector machine, graphical models, etc. Working on real and/or simulated data through assignments, students will apply the knowledge learned and practice their skills in solving various biological and public health problems, such as sequence alignment, gene prediction, subtype identification and classification, and disease risk and prognosis prediction. Discussion on model assessment and selection are also included. Elementary knowledge of the use of statistical computing packages is needed.