Course Syllabus

Jump to Today

ECON 310/STAT 376: Econometrics — Fall 2026

Professor: Mahmoud El-Gamal

Class: MW, 2:00—3:15 p.m., TBA

Lab: M, 5:00—5:50 p.m., TBA

Office hours: By appointment

TA: TBA

Course Description: 

We study various regression methods for estimation and inference with special emphasis on (1) understanding mathematically when various methods work as desired and why, (2) conducting simple Monte Carlo simulations, and (3) replicating famous econometric studies using the authors’ original data and regression methods

Textbook (Chapter numbers noted on syllabus):

AI-Assisted Learning:

  • The legendary late Math professor Paul Halmos wrote that "For a student of mathematics to hear someone talk about mathematics does hardly any more good than for a student of swimming to hear someone talk about swimming." The same applies to econometrics. In this course, you will learn to do econometrics by doing econometrics in class
  • We will use Positron, a VSCode fork that has a somewhat similar layout to RStudio but allows you to use the full suite of AI agents available through GitHub Copilot (you get free access to Copilot Student as a verified student -- as of April 20, 2026, student access has been reduced but not eliminated, and new enrollments beyond the free Copilot tier are paused; we'll see what's available in August; also, if you are an adventurous group, we can explore multi-agentic work with Roo-Code later in the semester; and I am experimenting over the summer on training some free local LLMs that may fit on your machines based on my coding style and preferences for the course)
    • I will guide you during the first class and lab to set up your environment. You will need to bring your notebook computer to every class and lab
    • You will begin with "vibe coding" (giving natural-language instructions to AI agents to generate R code; trusted R packages will serve as our statistical substrate as we study the various econometric methods covered in the textbook). We will explore how the code works and computationally dissect the mathematical foundations for the various methods of estimation and inference
  • I have my lecture notes in Quarto markdown documents, reproducing and extending the results in the textbook, and adding illustrations from published research, which we will discuss in greater detail
  • Classes will be interactive: I will only share the slides without the R code that generated them, and guide you to work with AI tools to understand the underpinning math, and to replicate and build upon my lecture analysis
    • You will be allowed to discuss this work with your neighbors, but you have to write and upload your own Quarto and rendered PDF files to Canvas
    • This classwork will constitute your graded assignments. They will be graded lightly: 8/10 for any meaningful work, 9/10 for partial work in the right direction, and 10/10 for complete or nearly complete correct work
    • Classwork assignments will remain open on Canvas for a week in case you wish to improve your grade, after which I will upload the Quarto files that I used to generate my lecture slides
  • You will need to document your interaction with the AI in your Quarto document, explaining in your own words what you instructed the AI to do, why, and how. You will be graded on the problem-solving process, not just the final product
  • There will be three exam sessions during which you will be required to work alone with AI help, and only work finished during the allotted class time will be graded for correctness and completeness

Attendance Policy?

  • I am not formally requiring in-person attendance, except on exam days. Slides and problem sets for regular classes will be posted on Canvas a few minutes before class, and the work completed during that time will be due within 10 minutes of the end of class. Therefore, while they would lose the ability to discuss problems with their classmates or me, students who choose not to attend class will still have to complete the work at the same time, but they will not lose any points for not attending in person

Tentative Syllabus: 

  • Week 01 -- Aug. 24, 26: Projection & Conditional Expectations basics (Ch. 2) 
  • Week 02 -- Aug. 31, Sep. 2: Conditional Expectations & Projection  (Ch. 2)
  • Week 03 -- Sep. 9: Least Squares Regression & Best Linear Unbiased Estimator  (Chs. 2-3)
  • Week 04 -- Sep. 14, 16: BLUE, GLS, Heteroscedasticity-Consistent and Clustered SEs (Ch. 4)
  • Week 05 -- Sep. 21: Heteroscedasticity and Autocorrelation Consistent SEs + Inference (Chs. 5-7)
    • September 23: Exam 1
  • Week 06 -- Sep. 28, 30: Diff in Diff, Hypothesis testing & Wald Test (Chs. 18, 7-9)
  • Week 07 -- Oct. 05, 07: Inference, Restricted Estimation, Jackknife & Bootstrap (Chs. 9-10)
  • Week 08 -- Oct. 14: Many Variables and Instruments & 2SLS  (Chs. 11-12)
  • Week 09 -- Oct. 19, 21: Generalized Method of Moments (Ch. 13)
  • Week 10 -- Oct. 26: Limited Dependent Variables: Logit and Tobit  (Chs. 25, 27)
    • October 28: Exam 2
  • Week 11 -- Nov. 02, 04: Panel Data and Dynamic Panels (Ch. 17) 
  • Week 12 -- Nov. 09, 11: Time Series: ARIMA, Unit Roots  (Chs. 14, 16)
  • Week 13 -- Nov. 16, 18: Multivariate Time Series: VAR + SVAR, Cointegration & VECM (Chs. 15-16)
  • Week 14 -- Nov. 23: Nonparametric & Polynomial Regression (Chs. 19-20)
  • Week 15 -- Nov. 30: Quantile Regression (Ch. 24)
    • December 2: Exam 3

Grading:

  • Assignments: 50% (lowest three assignment scores will be dropped)
  • Exams: 50%

Special Accommodations:

  • If you have a documented need for special accommodation, you should tell me as soon as possible so that we may make arrangements to accommodate your needs

 

Course Summary:

Course Summary
Date Details Due