Feedback: We welcome student feedback regarding the course at any point. Please feel free to email us directly, or leave anonymous feedback for the instructional team by using our online form.

Overview of the course

• L1: Course outline, examples of time series data
• L2: Models for time series data, stationarity, autocorrelation, and basic theoretical constructs
• L3: Trend models and data wrangling
• L4: AR, MA models, causality, invertibility
• L5: ARMA models
• L6: Forecasting, metrics
• L7: Estimation
• L8: ARIMA/SARIMA
• L9-L10: Spectral methods
• L11-L13: State space models
• L14: Multivariate time series: VAR model, Granger causality
• L15: Continuous and non-real-valued time series
• L16: Nonlinear models: Switching state space models, XGBoost
• L17-L19: Deep Learning methods
• L20: Review

Prerequisites: basic course in Statistics at the level of 200

Course Requirements: Familiarity with a programming language.

Website: The primary web site for the class is stats207.stanford.edu. We will also use Canvas for class lecture recordings.

Discussion. We will use Ed to allow you to get help efficiently from both your classmates and the instructors (class Ed page). Please post your questions about the course material and course logistics to Ed so that everyone can benefit from the answer. We also highly encourage you to answer your classmates’ questions whenever possible – you will get extra practice with the material and receive feedback from the teaching team about your answers. As an added incentive, students who provide frequent, high-quality answers may receive extra credit on their final course grades. Additional guidelines about posting on Ed are available on the Ed page.

The VPTL office offers tutoring for this class by appointment, which can be scheduled here.

Assessment and grading

• Homework assignments (40%)
  • Start early!
  • HW 1 (20%), HW 2 (20%)
• Midterm take-home exam (25%)
  • Start early!
• Final project (35%)
  • Midway report (20%)
  • Project presentation (20%)
  • Final paper (60%)

Collaboration policy

• Every student must first attempt all problems individually
• You may discuss a homework assignment with up to two classmates
• Each student must write up his/her own solutions and code individually
• Explicitly name any collaborators at the top of the homework
• Please don’t search for answers on the web, Google, etc.
• please ask us if you are not sure if you can use a particular reference

###Regrade requests

• Must wait 24 hours before submitting a regrade request after grading has been published
• Subsequently, requests submitted via Gradescope within one week of grades being published

Grade disputes: Students must wait 24 hours after receiving a grade before they can dispute it, after which disputes must be received within 7 days of receipt of the grade. Grade disputes must be submitted to the instructional team on Gradescope.