In this project, you should work in groups of (about) 4 persons.

Instructions

Download

Data files for Task 1 - 2

Download.

Task 3

In this task, you should look into some identification method that has not been covered in detail in the course.

Here are some suggestions for methods you can take a closer look at:

Ridge Regression (RR)
Neural Networks (NN)
Instrumental Variables (IV)
Subspace (SS) (see Chapter 13 in the main textbook, contact Per to discuss project details)
Aktivslamprocess (Rekursive identifiering)

When your group has decided what you will do in Task 3, hand in EMSP2023: Choice of method in Task 3.

Optional task

You can do an optional task in order to get a higher grade in the course. If your group, or parts of your group, plans to do such an optional task then hand in EMSP: Choice of optional task as soon as possible. If you have questions about the different possible tasks, contact Per Mattsson.

Coaching/Supervision

You can ask questions about the project during computer labs.

For Task 1-Task 2 , the main responsible teacher is Sofia Ek.

Oral presentation

Each group gets 10 minutes to present their work, and after this there will be time for questions.
Everyone in the group must actively participate in the presentation.

Note that everyone will do Task 1 and Task 2. So, you do not have to give any background to these tasks, and can instead directly present and discuss your results.

For Task 3, you should give a brief description of the method you have studied without going into too many technical details. Also explain how you evaluated the method, and what your results are.

If you did an optional project, you should not spend much time at all on Task 1-Task 3 (it is okay to skip it completely). Instead concentrate on presenting your work on the optional task.