Those of you who want to get 9CP for this course will need to complete a final project.

Specifically, you may choose between two possible routes for this project.

Theoretical: Exploring Topics through Geometric or Probabilistic Formalization

For the theoretical route, you will be asked to write a report of 5-10 pages about a topic of your choice, in which you elaborate on the formalization from either a geometric or probabilistic perspective. Here, the main goal is for you to showcase your understanding of geometric or probabilistic perspectives, drawing upon the fundamental concepts and methodologies covered in this course.

For the topic, you may choose any statistical method that interests you, provided you get my approval first.

• Details and ressources

  For the theoretical project, you should write a minimum of 5 pages, more if appropriate. For this version, the content of the project is more important than the style. Specifically, you should make sure to fulfill the following points:

  1. It’s clear that you fully understand the topic and have expressed it clearly and formally (that means using mathematical notation!) in your writing.
  2. You’ve successfully connected the dots, starting from:
     • (a) “What is the goal?” and
     • (b) “What does my data look like and how do I model it?”, then continuing to
     • (c) “These are the assumptions that are being made for this method/concept,” and, finally, elaborate on
     • (d) “Here’s how I can interpret my results given the data and assumptions.”
  3. All sources are cited, including lecture notes and Wikipedia articles, which are perfectly valid references for this project.

  You may simply submit this project by sending me the pdf

Applied: Implementing a statistical algorithm in a minimalistic R-package

For the applied route, you will be asked to write your own little R-package. In it, you should implement an algorithm that is within the broader scope of the course.

As with the theoretical route, the exact choice of algorithm is up to you, provided you get my approval first.

• Details and ressources

  For the applied project, it is important that you adhere to best practices in developing an R package, as laid out in “R Packages” by Hadley Wickham and Jennifer Bryan. Feel free to also use other helpful resources online!

Specifically, make sure that:

• The file structure of your package is correct.
• You provide detailed documentation of all exported functions.
• You do not forget to write tests.
• Additionally, your project should contain at least a README file that explains what the package does, which algorithms you have implemented, and how they work. The README should also include demonstrations of the most important functions.

Alternatively to the last point, you may create one or more vignettes and even include them on a package website (which is very easy to create using GitHub) if you’d like.

This project should be submitted by sending me the link to the GitHub repository where it is located.

General Formalities

If you are unsure whether your project fulfills the requirements above, you are welcome to submit it to me for an assessment. I will do my best to provide feedback, but please feel free to remind me if you haven’t received a response after three days.

If I believe that certain aspects of your project submission need clarification or that you might deserve a better grade than I would initially assign based solely on your submission, I reserve the right to schedule a meeting to discuss your project with you before determining your final grade.