Schedule for System Identification

The course starts on Wednesday, January 13, 2016, at 13.15 in Algoritmen.

The schedule below is tentative. It will be fixed at the first meeting 13/1.

All lectures will be in Algoritmen except on March 9. Then it will be in Transformen

>

Date	Notes	Chapters	Problems
We 13/1 13.15	Overview	...	...
We 20/1 13.15	Nonlinear Models	Kap 1,2,3	2G.4, 2E.2, 2E3, 3G.2, 3E.4
*Th 28/1 15.15	Linear Models - Special Issues	Kap 4,5	4G.4, 4G.5, 4E.1, 4E.3, 5G.1, 5E.2
We 3/2 13.15.	Chapters 8 - 9	6,7	6G.1, 6G.3, 6G.5, 7G.1, 7G.4,7E.1, 7E.3
*Th 18/2 15.15	...	8,9	8G.2 8G.3, 8E.3, 9G.2, 9E.3, 9E.4
We 24/2 13.15	...	Data Discussion	..
We 2/3 13.15	Experiment Design	Kap 10,11	10G.3,10E.1,10E.3,11E.4,11D.1,11E.3
*We 9/3 13.15	In Transformen	12,13,14	13G.2,13E.1,13E.3,13E.5.14G.2
We 16/3 13.15	...	15,16,17, Summary	15E.1,16G.1,16E.2,16E.3
We 23/3 13.15	...	Reports on micro-projects; Questions	..

Part of the last meeting will be devoted to questions on the whole material.

The exam period is three days that you select yourself in the period from Easter to mid April

The organization of the lectures is as follows: The material of the chapters will be discussed during the first half of the lecture on the indicated day. It is assumed that the participants have read and digested the material before that and also have solved the indicated problems. The participants should be prepared to take active part in the discussion of the material. The second half of the lecture will be used to introduce next week's material.

It is recommended that the participants meet an additional time per week, without the lecturer, to discuss the problems and prepare the discussion for the next lecture.

An errata list for the textbook is available here

The course contains two projects to be performed in groups of 1-3 students: An identification problem with real data and a "micro-project".

Data tests

Working with real data sets will be an essential part of the course. the DAISY identification site

and work with it during the first few weeks. Preferably the work should be carried out using the System Identification Toolbox. (Matlab 8 or higher). If you need some exercises to get further familiar with this toolbox you can download this file

Project ideas:

1. Investigate what the MATLAB-package LOLIMOT can do for LOcal LInear MOdels. How does it compare with Tree - NLARX models?
2. Matlab has a "Global Optimization" Toolbox. Investigate how such techniques can be used to avoid problems with local minima in identification problems.
3. "Elastic Nets" are combinations of L1 and L2 regularization. The Statistics Toolbox has mfiles for elastic nets for linear regression. Investigate how that works for ARX identification problems.
4. Arch and Garch models (ARX models with varying noise level) have been very popular in Econometrics since the "Nobel Prize" in 2003. Investigate how those techniques will handle ARX-problems in system identification, compared to ignoring the noise variation.
5. Test simple methods to estimate Hammerstein-Wiener models (L--NL--L).
6. Investigate differences between ML and PEM when there is noise structure knowledge.