Ctrl+K

STATS305C: Methods for Applied Statistics

STATS305C: Methods for Applied Statistics#

Lecture: MW 1:30PM-2:50PM, 540-108

References#

Lectures will be a mix between chalk talk (particularly first half focused on multivariate analysis) and computing examples.

Teaching Team, Office Hours, Discussion Section#

Name

Email

Role

Office hours

Discussion section

Jonathan Taylor

jtaylo

Instructor

MW 3:00-4:00 in Sequoia Hall 137

Michael Salerno

mdsalerno

TA

TBD

Computing environment#

We will use R for most calculations. Examples will typically be in the form of jupyter notebooks.

Prerequisites#

Stats 305A and 305B, some matrix analysis.

Course description#

The first half (or slightly more) of the course we will focus on some classical multivariate analysis from Mardia, Kent, Bibby. Where appropriate, connections to more modern versions of the classical ideas will be made.

The second half of the course will consist of more modern multivariate methods, often lumped together as “unsupervised learning”. This part of the course will use material from ESL (Elements of Statistical Learning).

Topics covered may include (with some additional ones):

  • Multivariate Gaussian and undirected graphical models;

  • Multivariate inference: Hotelling’s \(T^2\), canonical correlations;

  • Matrix decomposition: principal components analysis, factor analysis, matrix completion, correspondance analysis;

  • Discriminant analysis and recent variants;

  • Clustering: hierarchical, \(k\)-means, bi-clustering.

  • Multidimensional scaling: spectral methods, t-SNE.

Evaluation#

  • 5 assignments (50%)

  • project oral presentation + writeup (20%)

  • final exam (30%)

Final exam#

Following the Stanford calendar: Monday June 9, 2025 @ 3:30PM-6:3PAM

Gradescope: here soon#

Ed forum: see Canvas#

Honor Code#

Violating the Honor Code is a serious offense, even when unintentional.

Prohibited Conduct:

  • Copying from another’s exam

  • Unpermitted collaboration

  • Representing another’s work as one’s own

Full Honor Code available at: Stanford Community Standards

Contents