Course positioning
STATS 202 is the core data-science course in the statistics department, taught by Robert Tibshirani and others. It covers the main supervised and unsupervised methods from a statistical perspective, emphasizing model selection, evaluation, and interpretability. It is more applied and inference-oriented than CS 229, and suits people interested in data science.
What it covers
- Regression: linear, polynomial, logistic, LDA
- Model evaluation: cross-validation, bootstrap, model selection
- Regularization: ridge, lasso, elastic net
- Nonlinear methods: splines, GAMs, trees, random forests, boosting
- SVMs: kernels, margins
- Neural networks and Transformers: basics and applications
- Unsupervised: PCA, k-means, hierarchical clustering
Prerequisites
Probability/statistics and linear algebra; programming (R/Python).
Assessment & workload
Assignments plus an exam/project. Check the current syllabus for details.
Who it's for
Data science, statistical analysis, and business analytics; ML learners who want a statistical perspective.
Study advice
It complements CS 229: 202 emphasizes model evaluation and statistical inference, 229 emphasizes algorithm derivations.
Related courses
Official course description
Overview of supervised learning, with a focus on regression and classification methods. Syllabus includes: linear and polynomial regression, logistic regression and linear discriminant analysis; cross-validation and the bootstrap, model selection and regularization methods (ridge and lasso); nonlinear models, splines and generalized additive models; tree-based methods, random forests and boosting; support-vector machines; Neural nets and transformers; Some unsupervised learning: principal components and clustering (k-means and hierarchical).