Data Science

This class will introduce students to a variety of computer algorithms and the statistical models they represent in order to perform prediction or classification. Models covered include: k-nearest neighbors, linear and logistic regression, principal component analysis, k-means and hierarchical clustering, trees, random forests and support vector machines. The class discusses the fundamentals of experimental design, model validation, optimization, and comparison. Other topics include: quantile regression, resampling methods and introduction to Bayesian statistics. Students will learn to prepare and use datasets from the natural and social sciences to formulate and answer research questions relevant to their own lives. This class includes substantial use of statistical software.