Statistics 151A: Linear Models
UC Berkeley, Fall 2024
Instructors
Instructor: Ryan Giordano
Office: 389 Evans Hall
Office hours:
Tuesdays 2-3pm
Fridays 10-11am
rgiordano@berkeley.edu
pronouns: He / him
GSI: Haodong Ling
Office: Evans 428
Office hours:
Tuesdays from 4:00-5:30 PM
Wednesdays from 11:00 AM-12:00 PM and 4:00-5:30 PM
haodong_ling@berkeley.edu
pronouns: He / him
Please reach out to Haodong if you need to be manually added to BCourses.
Course content
This website will contain lecture materials and assignments. Day-to-day announcements can be found in BCourses. Discussions can be found in ED. (See links above.)
Schedule
Lectures will be held Aug 28 2024 – Dec 05 2024 on Tuesday and Thursday, 12:30 pm – 2:00 pm, in Social Science 20.
Labs will be held on Wednesdays from 9:00 am – 11:00 am and 2:00pm – 4:00 pm in Evans 342.
(Link to official course calendar.)
Linear algebra review materials
Basic linear algebra is a serious prerequisite for this course. You can find a summary of useful review materials here.
R programming review materials
This course will be conducting in the R programming language, and basic proficiency is assumed. Here are some useful review materials:
Materials
Unelss otherwise noted, the primary materials for the course are the lecture notes, which will be posted to the course website in advance of class. The following textbooks are useful supplementary texts and are all freely available online:
- VDS: Veridical Data Science Yu, Barter
- LME: Linear Models and Extensions Ding
- ROS: Regression and other Stories Gelman, Hill, Vehtari
- ISL: An Introduction to Statistical Learning James, Witten, Hastie, Tibshirani
- RDS: R for Data Science, Wickham, Grolemund
- ETM: Econometric Theory and Methods Davidson, MacKinnon
I will additionally recommend optional readings from
- FPP: Statistics Freedman, Pisami, Purves.
Unfortunately this book is not available digitally through any official channels.
(Tentative) Course Calendar
The following schedule will almost certainly change.
| Date | Day of week | Event | Topic | Notes | Supplementary reading |
|---|---|---|---|---|---|
| Aug 29 | Thursday | Lecture 1 | Sample means as inference | ||
| Sep 3 | Tuesday | Lecture 2 | Sample means as loss minimization and projection | ||
| Sep 5 | Thursday | Lecture 3 | Inference and confounding with sample means | ROS 1.1-1.4 | |
| Sep 10 | Tuesday | Lecture 4 | Ames Housing data for inference | VDS 8.4 | |
| Sep 12 | Thursday | Lecture 5 | One-hot encoding and sample means | HW 1 due Friday Sep 13 | VDS 10.2, ISL 3.3.1 |
| Sep 17 | Tuesday | Lecture 6 | Multilinear regression as loss minimization | ETM 1.4-1.5, LME 3.1 | |
| Sep 19 | Thursday | Lecture 7 | Examples of the matrix form of linear regression | Quiz 1 in class. | |
| Sep 24 | Tuesday | Lecture 8 | Redundant regressors | ||
| Sep 26 | Thursday | Lecture 9 | Linear regression as projection | HW 2 due Monday Sep 30 | ETM 2.1-2.3, LME 3.1-3.3 |
| Oct 1 | Tuesday | Lecture 10 | Transformations of regressors | VDS 10.3, ROS 10.1-10.4 | |
| Oct 3 | Thursday | Lecture 11 | Transformations of responses | Quiz 2 in class. Guest lecturer. | ETM 2.4, LMS 7-8 |
| Oct 8 | Tuesday | Lecture 12 | Influence and Outliers | LME 11 | |
| Oct 10 | Thursday | Lecture 13 | Influence and Outliers | HW 3 due Friday Oct 11 | LME 12.2 |
| Oct 15 | Tuesday | Lecture 14 | The FWL theorem | LME 7 | |
| Oct 17 | Thursday | Lecture 15 | Stochastic assumptions on the residual | Quiz 3 in class. | ETM 3.1-3.3, ETM 4.1, ROS 4.5 |
| Oct 22 | Tuesday | Lecture 16 | Omitted variables in inference and prediction | LME 9.2 | |
| Oct 24 | Thursday | Lecture 17 | Regression to the mean | HW 4 due Monday Oct 28 | ETM 8.2, FPP 10.4 |
| Oct 29 | Tuesday | Lecture 18 | Confidence intervals and hypothesis testing | Guest lecturer. | ETM 4.1, ROS 4.5 |
| Oct 31 | Thursday | Lecture 19 | Coefficient tests under normality | Quiz 4 in class. | ETM 4.4-4.5 |
| Nov 5 | Tuesday | Lecture 20 | Uncertainty in the residual variance | ||
| Nov 7 | Thursday | Lecture 21 | Testing under machine learning assumptions | HW 5 due Friday Nov 8 | LME 12, LME 6, ETM 5.5 |
| Nov 12 | Tuesday | Lecture 22 | Variable selection and the F-test | ETM 4.4-4.5 | |
| Nov 14 | Thursday | Lecture 23 | Bias-variance tradeoff in prediction | Quiz 5 in class. | ISL 2.2 |
| Nov 19 | Tuesday | Lecture 24 | Ridge or L2 regression | ISL 6.2 | |
| Nov 21 | Thursday | Lecture 25 | LASSO or L1 regression | Project draft due Monday Nov 25 | ISL 6.2 |
| Nov 26 | Tuesday | Lecture 26 | Project consultation | ||
| Nov 27 | Thanksgiving | (no class) | |||
| Nov 28 | Thanksgiving | (no class) | |||
| Nov 29 | Thanksgiving | (no class) | |||
| Dec 3 | Tuesday | Lecture 27 | Project consultation | ||
| Dec 5 | Thursday | Lecture 28 | Project consultation | Project final due Friday Dec 6 |