ECON9104C – Topics in Econometrics
Schedule, syllabus and examination date
Course content
Subtitle spring 2021: Machine Learning in Economics
This course is unfortunately fully booked.
This course offers a comprehensive review of fundamental tools from machine learning (ML) and relevant applications of these.
Before the course participants are expected to have knowledge of a scripting language like Julia, Python, R or Stata. The course will be based mainly on Python using packages in R.
Learning outcome
Here is a tentative schedule for course outline:
- Session 1: introduction, prediction policy problems, prediction quality, regularization
- Session 2: pipeline, gradient descent, generalization error, cross-validation
- Session 3: tree-based models e.g. random forest, kernel models e.g. nearest neighbor
- Session 4a: nested cross-validation, working with text-data
- Session 4b: flashtalks on research ideas
- Session 5: ML in estimation 1 - applications in regression and instrumental variables (post-Lasso and applications)
- Session 6: ML in estimation 2 - applications in matching (causal forest, generalized random forest)
- Session 7a: unsupervised learning (e.g. k-means, principal components)
- Session 7b: presentation of research ideas
Admission
This course is offered to PhD candidates at the Oslo PhD Initiative in Economics at UiO and BI. Other candidates admitted to a PhD program may apply.
Teaching
The course takes place during one intensive week.
Credit for the course requires both active participation in class and a take home exam.
Physical classroom instruction is planned for this course.
However, if, due to the infection situation, we cannot have physical teaching, the course will be given either as a hybrid or fully digital on zoom.
If the course ends up going digital on zoom, the format will have to be changed somewhat.
Examination
Exam in two parts:
- Multiple choice.
- Exam paper that outlines a research design using ML in economics.
Language of examination
The examination text is given in English, and you submit your response in English.
Grading scale
Grades are awarded on a pass/fail scale. Read more about the grading system.