Schedule

Preliminary schedule

Homeworks within parentheses are additional problems for FEL3202.¶


* Introduction (Friday 15/5, 15-17) . Chapter 1-2 in Lecture Notes (LN). Chapter 1-2 in Ljung.
* The basic problem
* Some examples
* Model selection using ranking
* Some pitfalls
* HW: 1.1 a-d (1.1f)

* Probabilistic models (Tuesday 19/5, 10-12). Chapter 3 in LN. Chapter 4 in Ljung.
* Models and model structures
* Estimators
* A probabilistic toolshed

* Estimation theory and Wold decomposition (Tuesday 26/5, 10-12). Chapter 4 in LN. Chapter 3 in Ljung
* Estimation theory
* Information contents in random variables
* Estimation of random variables

* Wold decomposition

* Unbiased parameter estimation (Friday 29/5, 15-17). Chapter 5 in LN. Chapter 7 in Ljung.
* The Cramér-Rao lower bound
* Efficient estimators
* The maximum likelihood estimator
* Data compression
* Uniform minimum variance unbiased estimators
* Best linear unbiased estimator (BLUE)
* Using estimation for parameter estimation

* Biased parameter estimation (Tuesday 2/6, 10-12) . Chapter 6 in LN.
* The bias-variance trade-off
* The Cramér-Rao lower bound
* Average risk minimization
* Minimax estimation
* Pointwise risk minimization

* Asymptotic theory (Friday 5/6, 15-17). Chapter 7 in L.N. Chapter 8
* Limits of random variables
* Large sample properties of estimators
* Using estimation for parameter estimation, part II
* Large sample properties of biased estimators

* Computational aspects (Tuesday 9/6, 08-10). Chapter 10 in Ljung.
* Gradient based optimization
* Convex relaxations
* Integration by Markov Chain Monte Carlo (MCMC) methods

* Case studies I (Friday 12/6, 10-12)
* Parametric LTI models
* Impulse response models

* Case studies II (Tuesday 16/6, 10-12)
* Uncertain input models
* Nonlinear stochastic state-space models

* Model accuracy (Friday 19/6, 15-17) Chapter 9 in Ljung.
* Model structure selection and model validation (Tuesday 23/6, 10-12). Chapter 16 in Ljung
* Experiment design (Tuesday 25/8, 10-12) . Chapter 13 in Ljung.
* Continuous time identification (Friday 28/8, 15-17)