Course Syllabus
Contact Information
Welcome to class! I am Dr. Tyler Ransom, and I designed and am teaching this course on Machine Learning in Python for economists. I earned my PhD in economics from Duke University in 2015 after completing my undergraduate degree in economics at Brigham Young University in 2009. I have been at OU's economics department since 2017.
My academic research covers a variety of topics, including labor economics, economics of education, urban & regional economics, applied econometrics, and health economics. You can learn more about my previous and ongoing work by visiting my personal website or Google Scholar profile page.
Course Details
For a list of course activities, scroll to the bottom of this page.
Course Delivery
This course is multi-modal: learning can be entirely asynchronous since Live Session attendance is not required, but synchronous Live Sessions on Zoom will be offered. Otherwise, everything will be asynchronous, though there will be deadlines.
Course Prerequisites
The course is open to all students in the program, with no prerequisites required. However, having a basic understanding of college algebra, computer skills, and some quantitative experience would be advantageous.
Course Materials
Required:
- James, G., Witten, D, Hastie, T., & Tibshirani, R. (2023). An introduction to statistical learning with applications in Python. Springer. An Introduction to Statistical Learning.
- Python coding environment: Google Colab
Program Learning Outcomes (PLO)
-
Analysis and Insight: Use econometrics models to analyze complex economic phenomena, providing valuable insights for decision-making and policy formulation in various domains, such as economics, finance, and social sciences.
-
Application & Data-driven Decisions: Use advanced econometric methods to solve real-world economic problems and make evidence-based decisions across various domains based on advanced statistical models, to improve outcomes for individuals and society.
-
Data Visualization: Convert intricate data into easily comprehensible visuals for effective data analysis, improving decision-making, and communicating insights to mass audiences.
-
Effective Communication of Research Findings: Effectively communicate research findings through written reports, presentations, and visuals to ensure that both specialized audiences and the general public understand the research.
Course Learning Outcomes (CLO) Alignment
| CLO | Description | PLO |
|---|---|---|
| A |
Implement machine learning algorithms using Python |
1, 2 |
| B |
Evaluate and compare performance of different ML models |
1, 2 |
| C |
Analyze real-world applications of ML in economics |
2, 4 |
| D |
Interpret and communicate results of ML analyses |
3, 4 |
| E |
Understand how ML (prediction) and econometrics (causality) complement each other |
1, 2 |
| F |
Design ML solutions for specific economic problems |
1, 2, 3 |
Grades
Breakdown
| Activity | Description | Points/Percentage |
|---|---|---|
| Quizzes & Participation | Quizzes will happen periodically and will be open book. They will be 10 questions and due midweek. There will be 6 quizzes at 100 points each. There will also be other required participation activities that will count towards this part of your grade. | 10% |
| Problem Sets | Periodically, you will have an in-depth homework assignment consisting of a problem set of 5-10 open-ended questions. There will be 6 problem sets at 100 points each. | 30% |
| Project Check-ins | Throughout the semester, you will work on your final project one step at a time and get feedback along the way. There will be 3 project checkpoints at 100 points each. | 30% |
| Final Project | At the end of the semester, you will submit your completed final project, incorporating the feedback you have gotten along the way. The final project will be worth 100 points. | 30% |
| Total | 100% |
Scale
Grades will be given as earned; there will be no grading on a curve.
| Percentage | Letter Grade |
|---|---|
| 90–100% | A |
| 80–89% | B |
| 70–79% | C |
| 60–69% | D |
| Below 60% | F |
Course Components
Module Introduction/Overview Pages
Each module will have an introduction/overview page, which will provide background or context to the topic, outline specific skills/competencies you are expected to gain from actively engaging with the module, and provide a checklist to help you manage your time.
Resources
Each module's resources page will offer foundational information you need to be successful in the given module. It may include readings, instructional videos, etc. Please ensure you complete the readings and watch the instructional videos before the Live Session.
Quizzes
About once every two weeks, you will have a quiz. Quizzes will be due on Mondays at 6 PM (Central) so that the instructor will have the chance to be responsive to the needs of the class prior to the Live Session. Quizzes will be composed of multiple choice questions designed to test your understanding and retention of the assigned reading & lecture material. They will also help reinforce key concepts, which will prepare you for the upcoming modules, where you will be asked to apply or extend their learning. These multiple-choice quizzes will be short and to the point, ensuring that you stay engaged with the course content and must be completed before Live Sessions. These will be auto-graded and help you to judge your comprehension of key concepts. You will get the chance to take these quizzes multiple times, as they are intended to be a low-stakes learning opportunity. If you do poorly the first time, that is a good sign that you need to review all of the module materials in more depth.
Live Sessions
Every two weeks, we will have an optional synchronous Live Session on Zoom in which there will be time for questions, workshops, and interactive discussion of course concepts. These will offer important opportunities to interface with your instructor and peers.
Problem Sets
About once every two weeks, you will have a problem set that mainly covers all concepts since the previous problem set. Problem sets will be due on Sundays at 11:59 PM (Central). These problem sets will help you and me to gauge your progress as well as where you need to grow to succeed in the remainder of the course. Problem sets will include about 5-10 analytical open-ended questions and real data analysis assignments using Python. You will be required to submit all code used in generating your results. You can find the Problem Sets Rubric attached to each problem set in the modules. These should be submitted in a professional format such as PDF.
Final Project & Project Checkpoints
The project checkpoints will offer opportunities for feedback as you complete various parts of the final project. Project checkpoint assignments will be due on Sundays at 11:59 PM (Central). Additionally, I will provide brief guidance emails on Canvas ahead of each project step deadline. The final project report will be a summative assessment of your mastery over the course materials in an applied scenario. The submission deadline for final project report is the last day of finals week at 11:59 PM (Central). Please also ensure you watch the project overview video for more information about the research project. You can also find the generic rubric for the project under the Rubrics area.
Course Policies
Communication
The best way to contact me is via email at ransom@ou.edu. Please address me as Dr. Ransom or Prof. Ransom. I will respond to emails in the order received within 48 hours; emails sent over the weekend will be answered by 11:59 PM the following Tuesday (Central Time).
As your instructor, I am here to guide and support you, and I encourage you to reach out via email or during office hours (schedule by email) with any questions. However, keep in mind that, to get the most out of this course, you need to come prepared to engage in Live Sessions, which requires you to complete the readings and videos beforehand. The same preparation applies to your visits to my office hours. Remember, your contributions are valuable, and open communication is key to our collective success.
Feedback
Quizzes will be graded automatically. You can expect the problem sets and project steps to be graded within 7 days. Solutions for quizzes and problem sets will be available the day after their due dates.
Late Policy
Late quizzes will not be accepted, resulting in an automatic zero. Ensure you complete quizzes well in advance of the deadline. Only documented excuses will be considered for problem sets, project steps, and the final project report.
Plagiarism
Plagiarism is the most common form of academic misconduct at OU. There is no college-level assignment that can be satisfactorily completed by copying. OU's basic assumption about writing is that all written assignments show the student's own understanding in the student's own words. That means all writing assignments, in class or out, are assumed to be composed entirely of words generated (not simply found) by the student, except where words written by someone else are specifically marked as such with proper citation. Including other people's words in your paper is helpful when you do it honestly and correctly. When you don't, it's plagiarism.
For more information about plagiarism, watch this video and then consider taking this short course offered by University Libraries.
Generative AI Policy
Based on the Assignment
Throughout this course, you will develop the skills needed to effectively use Generative AI as an aid in your learning and in preparation for our changing field. There will be times in which you will not be permitted to use Generative AI or may only use it in particular ways. These decisions are intentionally made to support you in developing the skills and content knowledge needed in order to effectively use Generative AI. Follow assignment instructions carefully, as they will guide you in what you are permitted to use Generative AI for in each assignment. Where Generative AI is used, you must follow assignment instructions for appropriate citation and reflection about your usage.
Use of Generative AI outside of the scope of what is explicitly defined in our assignments, and without acknowledgment, will be considered a violation of the academic integrity policy for this course.
If you have any questions about how Generative AI may or may not be used on an assignment, please talk with me.
Consequences for Violating the Generative AI Usage Policy: Deviating from guidelines provided in each assignment may be considered a violation of the academic integrity policy of this course. Per our usage policy, you will be responsible for accuracy, including appropriately citing and summarizing any articles you find through AI research tools, and thus must read the material you are citing. Submitting data or research that is not real (a risk when overly relying on Generative AI) may result in an academic integrity violation for falsifying information. Additionally, there may be times, such as in-class quizzes, midterms, or finals, where Generative AI usage is prohibited. Any use of AI in those cases will be considered a violation of the academic integrity policy.
Note: For auto-graded quizzes, you may use any resources including AI to help you learn the material, as these are designed as low-stakes learning opportunities where multiple attempts are allowed.
My Use of Generative AI: I will model appropriate Generative AI usage by clearly disclosing when I use it and why. Expected use cases include: using AI to assist in providing feedback on assignments and suggesting preliminary grades (though all final grading decisions remain mine), creating course materials, revising assignment instructions and rubrics to improve clarity for students, and using AI research tools to find current articles. For the final project, to maintain consistency with the independent work expected of students, I will not use AI in the grading process.
University Academic Policies and Student Support
Access the University Academic Policies Document.
Course Summary:
| Date | Details | Due |
|---|---|---|