Nyhetsflöde

Fredag 229 november 2013 kl 10:15 - 12:00

A Minimum-Mean-Square Error Perspective on Model Selection
* Model selection as the minimization of the mean-square estimation error
* SURE - Stein's Unbiased Risk Estimate
* ML-estimation
Reading material Optional material  Lecturer Håkan Hjalmarsson¶

Fredag 29 november 2013 kl 10:15 - 121:00

Selected Topics 45 (FEL3202 Extended Course (12 credits))

K53

Friday 29 NovTuesday 3 December 2013 at 103:15 - 114:00

K53

L31

Selected Topics 43 (FEL3202 Extended Course (12 credits))

FreTisdag 159 november 2013 kl 10:15 - 11:00

Subspace identificationFundamental properties ¶

Optional Reading ¶

Lecturer Cristian Rojas¶

Fundamental properties

Optional Reading ¶Lecture Slides
*
Lecturer Cristian Rojas

Fundamental properties

* Asymptotic covariance expressions (for high model orders)
* Proof of a.s. convergence results
Lecture Slides
* Special topics 3
Lecturer Cristian Rojas

Experiment Design and Closed-Loop Identification Bias/variance issues, experiment design, closed-loopSystem Estimation Methods III: Subspace identification¶ Reading Material Ljung: Chapters 12, 13¶ Optional References Slides 8, ¶ Slides 9¶ 4.3, 7.3, 10.6¶

Optional Material Lecturer Cristian Rojas

System Estimation Methods III: Subspace identification Reading Material Ljung: Chapters  4.3, 7.3, 10.6

Optional Material
* Slides 4
* Slides 5
* Max-Max problem
Lecturer Cristian Rojas

Lecture 6Student presentations 1

Torsdag 14 november 2013 kl 101:15 - 12:00

System Estimation Methods III: Subspace identification Reading Material Ljung: Chapters  4.3, 7.3, 10.6¶ Optional Material
* Slides 4
* Slides 5
* Max-Max problem
Lecturer Cristian Rojatudent Presentations
* Bayesian Inference¶


* Inconsistency of the maximum likelihood method¶



* Kernel methods

Student Presentations
*  Bayesian Iinference¶
* Inconsistency of the maximum likelihood method

* Kernel methods

Student Presentations
*  Bayesian inference
* : Inconsistency of the mMaximum lLikelihood method¶
Estimation
* Kernel methods

Student Presentations
*   Bayesian inference
* : Inconsistency of Maximum Likelihood Estimation
* Kernel methods

Student Presentations
*   Bayesian inference
* : Inconsistency of Maximum Likelihood Estimation
*  Non-parametric identification:Kernel methods
*

Student Presentations
*   Bayesian inference
* : Inconsistency of Maximum Likelihood Estimation
*  Non-parametric identification:Kernel methods
*

Student Presentations
*   Bayesian inference
* : Inconsistency of Maximum Likelihood Estimation
*  Non-parametric identification:Kernel methods
Lecturer Håkan Hjalmarsson¶

Torsdag 14 november 2013 kl 11:1500 - 12:00

Q13

Student Presentations
*   Bayesian inference
* : Inconsistency of Maximum Likelihood Estimation
*  Non-parametric identification:Kernel methods
Lecturer Håkan Hjalmarsson

Student Presentations
* Bayesian inference
* Inconsistency of Maximum Likelihood Estimation
* Non-parametric identification: Kernel methods
Lecturer Håkan Hjalmarsson

Problem solving session: Exercise Set 2

, Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   3E1, 3E2, C2a--d      

FEL 3202: 3T2, C2e-g.

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   3E1, 3E2, C2a--d   , 6G3   

FEL 3202: 3T2, C2e-g, 6T2

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   3E1, 3E2, C2a--d, 6G3   

FEL 3202: 3T2, C2e-g, 6T2

Lecturer Håkan Hjalmarsson¶

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   3E1, 3E2, C2a--d, 6G3   

FEL 3202: 3T2, C2e-g, 6T2

Lecturer Håkan Hjalmarsson

L43

Problem solving session: Exercise Set 4

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   7E2, 7G1, 8E2¶

FEL 3202:  7E8, 8E4, 8G5¶

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   7E2, 7G1, 8E2, 9E3

FEL 3202:  7E8, 8E4, 8G5

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   7E2, 7G1, 8E2, 9E3

FEL 3202:  7E8, 8E4, 8G5

Lecturer Cristian Rojas¶

K53

Friday 29 NovTuesday 3 December 2013 at 114:15 - 125:00

K53

L31

Student presentations
* Errors-in-variables
*  Structured estimation
* 1: Low rank  
* Nuclear normapproximation using local search
*  Structured estimation 2: Nuclear norm minimization

Student presentations
*
*  Structured estimation 1: Low rank approximation using local search
*  Structured estimation 2: Nuclear norm minimization
*  Errors-in-variables identification

Student presentations
*  Structured estimation 1: Low rank approximation using local search
*  Structured estimation 2: Nuclear norm minimization
*  Errors-in-variables identification

Student presentations
*  Structured estimation 1: Low rank approximation using local search
*  Structured estimation 2: Nuclear norm minimization
*  Errors-in-variables identification
Lecturer Cristian Rojas¶

Fredag 22 november 2013 kl 11:1500 - 12:00

Fredag 22 november 2013 kl 11:0015 - 12:0015

Model Quality Evaluation
* A geometric framework
 ¶

Model Quality Evaluation
* A geometric framework
 Reading material
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
* Variance Analysis in SISO Linear Systems, J. Mårtensson and H. Hjalmarsson, Draft.
¶

Optional material  Lecturer Håkan Hjalmarsson

Model Quality Evaluation
* Stochastic convergence concepts
* A geometric framework
Reading material
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
* Variance Analysis in SISO Linear Systems, J. Mårtensson and H. Hjalmarsson, Draft.

Optional material  Lecturer Håkan Hjalmarsson

Model Quality Evaluation
* Stochastic convergence concepts
* Stein's paradox
* Superefficiency
* Other bounds than the Cramér-Rao
* A geometric framework
Reading material
* Slides      8
* Slides 9
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
* Variance Analysis in SISO Linear Systems, J. Mårtensson and H. Hjalmarsson, Draft.

Optional material  Lecturer Håkan Hjalmarsson

Model Quality Evaluation
* Stochastic convergence concepts
* Stein's paradox
* Superefficiency
* Other bounds than the Cramér-Rao
* A geometric framework
Reading material
* Slides      8
* Slides 9
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
* Variance Analysis in SISO Linear Systems, J. Mårtensson and H. Hjalmarsson, Draft.

Optional material  Lecturer Håkan Hjalmarsson

Model Quality Evaluation
* Stochastic convergence concepts
* Stein's paradox
* Superefficiency
* Other bounds than the Cramér-Rao
* A geometric framework
Reading material
* Slides 8
* Slides 9
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
* Variance Analysis in SISO Linear Systems, J. Mårtensson and H. Hjalmarsson, Draft.

Optional material  Lecturer Håkan HjalmarssonCristian Rojas

Selected Topics 34 (FEL3202 Extended Course (12 credits))

L43

Model Quality Evaluation
* Stochastic convergence concepts
* Stein's paradox
* Superefficiency
* Other bounds than the Cramér-Rao
* A geometric framework
Reading material
* Slides 8
* Slides 9
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
* Variance Analysis in SISO Linear Systems, J. Mårtensson and H. Hjalmarsson, Draft.
Lecture Slides
*

Optional material   Lecturer Cristian Rojas

Model Quality Evaluation
* Stochastic convergence concepts
* Stein's paradox
* Superefficiency
* Other bounds than the Cramér-Rao
* A geometric framework
Reading material
* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
Lecture Slides
* Decay rate of the Cramér-Rao bound
* Barankin bound and SNR-threshold effect
* Superefficiency
* Nuisance parameters
Reading material Lecture Slides
* Special topics 4
Lecturer Cristian Rojas

FreTorsdag 154 november 2013 kl 10:15 - 11:00

Problems (should have been handed in at previous lecture) FEL 3201 (8 credits) + FEL 3202(12 credits):   4G4, 4E1, 4E5a, C3       

FEL 3202:  4E5b, C4

Lecturer Cristian Rojas¶

Q13

Information contents in data
* Sufficient statistics
* Rao-Blackwells theorem
Reading material E.L Lehmann: Theory of Point Estimation¶

Optional material.  Lecturer Håkan Hjalmarsson¶

Information contents in data
* Sufficient statistics
* Rao-Blackwells theorem
* Exact and conditional ML
Reading material E.L Lehmann: Theory of Point Estimation

Optional material.
* Slides 3
 Lecturer Håkan Hjalmarsson

L31

Information contents in data
* Sufficient statistics
* Rao-Blackwells theorem
* Exact and conditional ML
Reading material E.L Lehmann: Theory of Point Estimation

Optional material
* Slides 3 (old)
 Lecturer Håkan Hjalmarsson

Problems (should have been handed in previous lecture) FEL3201 & FEL 3202: 2G3, 2E1, 2E2, 2E5      

FEL3202: 2E3,  2T4, 2D3, 2D4

Problems (should have been handed in at previous lecture) FEL3201 & FEL 3202: 2G3, 2E1, 2E2, 2E5      

FEL3202: 2E3,  2T4, 2D3, 2D4

Problems (should have been handed in at previous lecture) FEL3201 & FEL 3202: 2G3, 2E1, 2E2, 2E5, 3E1, 3E2, C2a--d      

FEL3202: 2E3,  2T4, 2D3, 2D4, 3T2, C2e-g.

Problems (should have been handed in at previous lecture) FEL3201 & FEL 3202: 2E1, 2E2, 2E5, 3E1, 3E2, C2a--d      

FEL3202: 2E3,  2T4, 2D3, 2D4, 3T2, C2e-g.

Problems (should have been handed in at previous lecture) FEL3201 & FEL 3202: 2E1, 2E2, 2E5

FEL3202: 2E3,  2T4, 2D3, 2D4,

Lecturer Håkan Hjalmarsson¶

M35

Lecture 154

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Lecture 14f

MånTisdag 27 januari4 december 20143 kl 10:1501:00 - 102:00

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶ Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Lecture 2-ASelected Topics 1

Signal Estimation - Selected Topics
* Some concepts from probability theory
* Geometrical interpretation of estimation
* Properties of the Kalman filter
Reading material C.E. De Souza, M.R. Gevers and G.C. Goodwin: Riccati equations in optimal filtering of nonstabilizable systems having singular state transition matrices, IEEE Transactions on Automatic Control, 31(9), opp. 831-838¶

Optional material -¶

Lecturer Håkan Hjalmarsson¶

Signal Estimation - Selected Topics
* Some concepts from probability theory
* Geometrical interpretation of estimation
* Properties of the Kalman filter
Reading material C.E. De Souza, M.R. Gevers and G.C. Goodwin: Riccati equations in optimal filtering of nonstabilizable systems having singular state transition matrices, IEEE Transactions on Automatic Control, 31(9), opp. 831-838

Optional material -

Lecturer Håkan Hjalmarsson

Signal Estimation - Selected Topics
* Some concepts from probability theory
* Geometrical interpretation of estimation
* Properties of the Kalman filter
Reading material
* C.E. De Souza, M.R. Gevers and G.C. Goodwin: Riccati equations in optimal filtering of nonstabilizable systems having singular state transition matrices, IEEE Transactions on Automatic Control, 31(9), pp. 831-838

* probability_theory_1.pdf
Optional material -

Lecturer Håkan Hjalmarsson

Selected Topics 1 (FEL3202 Extended Course (12 credits))

B23

Asymptotic Properties Consistency,   convergence and asymptotic normality of estimation methods.

Reading Material Ljung: Chapters 8,9

Optional references Slides 6,

Slides 7

Lecturer Cristian Rojas¶

Asymptotic Properties Consistency,   convergence and asymptotic normality of estimation methods.System Estimation Methods II: Structured Estimation Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapters 8,9 7 (except 7.3)

Optional references
* Slides 6, ¶ Slides 7¶ 1  
* Slides 2
* Slides 3
* Slides 4
* Slides 5
* Max-Max problem
Lecturer Cristian Rojas

System Estimation Methods II: Structured Estimation Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapter 7 (except 7.3)

Optional referencesMaterial
* Slides 1  
* Slides 2
* Slides 3
* Slides 4
* Slides 5
* Max-Max problem
Lecturer Cristian Rojas

System Estimation Methods II: Structured Estimation Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapter 7 (except 7.3)

Optional Material
* Slides 1  
* Slides 2
* Slides 3
* Slides 4
* Slides 5
* Max-Max problem
Lecturer Cristian Rojas

System Estimation Methods III: Structured Estimation using Subspace identification Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapter 7 (except 7.3)

Optional Material
* Slides 1  
* Slides 2
* Slides 3
Lecturer Cristian Rojas

System Estimation Methods III: Structured Estimation using Subspace identification Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapter 7 (except 7.3)s  4.3, 7.3, 10.6

Optional Material
* Slides 1  
* Slides 2
* Slides 3
Lecturer Cristian Rojas

System Estimation Methods III: Structured Estimation using Subspace identification Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapters  4.3, 7.3, 10.6

Optional Material
* Slides 1  
* Slides 2
* Slides 3      4
Lecturer Cristian Rojas

System Estimation Methods III: Structured Estimation using Subspace identification Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapters  4.3, 7.3, 10.6

Optional Material
* Slides      4
Lecturer Cristian Rojas

L31

System Estimation Methods III: Structured Estimation using Subspace identification Model structures, Maximum likelihood estimation, The prediction error method,  Instrumental variables

Reading Material Ljung: Chapters  4.3, 7.3, 10.6

Optional Material
* Slides 4
Lecturer Cristian Rojas

System Estimation Methods III: Structured Estimation using Subspace identification

Reading Material Ljung: Chapters  4.3, 7.3, 10.6

Optional Material
*Lectur Slides 4
*  
Lecturer Cristian Rojas

System Estimation Methods III: Structured Estimation using Subspace identification

Reading Material Ljung: Chapters  4.3, 7.3, 10.6

Lectur Slides
*  Lecture 5
Lecturer Cristian Rojas

Parametric Estimation Methods Least squares, prediction error methods, instrumental variables and subspace methods.

Reading Material Ljung: Chapters 7, 10.6

Optional References Ljung, 1978

Slides      1,

Slides      2,

Slides      3,

Slides      4,

Slides      5,

Max-Max      problem,

Lecturer Håkan Hjalmarsson¶

ParametricSystem Estimation  Methods Least squares, prediction error methods, instrumental variables and subspace methods.¶ Reading Material Ljung: Chapters 7, 10.6¶ Optional References Ljung, 1978¶ Slides      1,¶ Slides      2,¶ Slides      3, ¶ Slides      4, ¶ Slides      5,¶ Max-Max problem,I: Basic principles and unstructured estimation Basic principles for matching models and data, Non-parametric identification¶

Reading Material Ljung: Chapters 4.1-4.2, 4.5, , 6¶

Optional Material L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.

Lecturer Håkan Hjalmarsson

System Estimation Methods II: Basic principles and unsStructured eEstimation Basic principles for matching models and data, Non-parametric identification

Reading Material Ljung: Chapters 4.1-4.2, 4.5, , 6

Optional Material L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.

Lecturer Håkan Hjalmarsson

System Estimation Methods II: Structured Estimation Basic principles for matching models and data, Non-parametric identification¶
* Model structures
* The prediction error method
* The instrumental variables method
Reading Material Ljung: Chapters 4.1-4.2, 4.5, , 6

Optional Material L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.

Lecturer Håkan Hjalmarsson

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.23, 47.5, , 6-7.7

Optional Material L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.

Lecturer Håkan Hjalmarsson

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Optional Material L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.¶ Lecturer Håkan Hjalmarssonecturer Cristian Rojas

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Optional Material
* Slides 1  
* Slides 2
Lecturer Cristian Rojas

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Optional Material
* Slides 1  
* Slides 2
Lecturer Cristian Rojas

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Optional MaterialLecture Slides
* Slides 2
Lecturer Cristian Rojas

Q11

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Lecture Slides
* Slides 2
Lecturer Cristian Rojas

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Lecture Slides
* Lecture 4
Lecturer Cristian Rojas

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Lecture Slides
* Lecture 4
Lecturer Cristian Rojas

System Estimation Methods II: Basic Principles cont'd, Structured Estimation
* Model structures
* The prediction error method
* The instrumental variables method
* Frequency domain identification
Reading Material Ljung: Chapters 4.1-4.3, 7.5-7.7

Lecture Slides
* Lecture 4
Optional Material J. C. Agüero, J. I. Yuz, G. C. Goodwin, R. A. Delgado, On the equivalence of time and frequency domain maximum likelihood estimation, Automatica, 46(2), 2010, 260-270.¶

Lecturer Cristian Rojas

Models and Nonparametric  Methods Models of linear time invariant systems. Nonparametric time- and frequency-domain methods.

Reading material Ljung: Chapters 4, 6

Optional references Lemma 2.1

Ljung,1985

Models and Nonparametric  Methods Models of linear time invariant systems. Nonparametric time- and frequency-domain methods.

Reading material Ljung: Chapters 4, 6

Optional references Lemma 2.1

Ljung,1985

Lecturer Håkan Hjalmarsson¶

Models and Nonparametric  Methods Models of linear time invariSignal Estimation Methods Prediction, filtering antd systems. Nonparametric time- and frequency-domain methodsmoothing. Optimal  filtering. Orthogonality condition. Wiener/Kalman/particle filter.

Reading material Ljung: Chapters 4, 63

Optional references Lemma 2.1

Ljung,1985¶ Lecturer Håkan Hjalmarsson

Signal Estimation Methods Prediction, filtering and smoothing. Optimal  filtering. Orthogonality condition. Wiener/Kalman/particle filter.

Reading material Ljung: Chapters 3

Optional references Lemma 2.1

Lecturer Håkan Hjalmarsson

Signal Estimation Methods Prediction, filtering and smoothing. Optimal  filtering. Orthogonality condition. Wiener/Kalman/particle filter.

Reading material Ljung: Chapter 3

Optional referencesmaterial Lemma 2.1

Lecturer Håkan Hjalmarsson

Signalystem Estimation Methods Prediction, filtering and smoothing. Optimal  filtering. Orthogonality condition. Wiener/Kalman/particle filterI: Basic principles and unstructured estimation
* .

Reading material Ljung: Chapter 32.4, 6,

Optional material Lemma 2.1

Lecturer Håkan Hjalmarsson

System Estimation Methods I: Basic principles and unstructured estimation
* .Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* .The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6,

Optional material Lemma 2.1

Lecturer Håkan Hjalmarsson

System Estimation Methods I: Basic principles and unstructured estimation
* Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* .The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6, 7.1-7.3 (not 208-211), 7.4

Optional material Lemma 2.1

Lecturer Håkan Hjalmarsson

System Estimation Methods I: Basic principles and unstructured estimation
* Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* .The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6, 7.1-7.3 (not pp. 208-211), 7.4

Optional material Lemma 2.1

Lecturer Håkan Hjalmarsson

System Estimation Methods I: Basic principles and unstructured estimation
* Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* .The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6, 7.1-7.3 (not pp. 208-211), 7.4

Optional material Lemma 2.1. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.

Lecturer Håkan Hjalmarsson

* System Estimation Methods I: Basic principles and unstructured estimation
* Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6, 7.1-7.3 (not pp. 208-211), 7.4

Optional material
* L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.

* Slides 1  
Lecturer Håkan Hjalmarsson

* System Estimation Methods I: Basic principles and unstructured estimation
* Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6, 7.1-7.3 (not pp. 208-211), 7.4

Optional material
* L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.
* Slides 1  
* Lemma 2.1
Lecturer Håkan Hjalmarsson

* System Estimation Methods I: Basic principles and unstructured estimation
* Non-parametric estimation
* Least-squares estimation
* Prediction error minimization
* The maximum likelihood method
Reading material Ljung: Chapter 2.4, 6, 7.1-7.3 (not pp. 208-211), 7.4

Optional material
* L. Ljung.     “On the Estimation of Transfer Functions”. Automatica, Vol. 21(6), pp. 677-696, 1985.
* Slides 1  
* Lemma 2.1
Lecturer Håkan Hjalmarsson

V12

Signals and Systems Linear time invariant systems and their uses (prediction, simulation).

Reading material Ljung: Chap 2,3

Optional references Lecturer Håkan Hjalmarsson¶

Signals and Systems Linear time invariant systems and their uses (prediction, simulation).Stationary stochastic processes. Quasi-stationarity. Convergence properties

Reading material Ljung: Chap 2,3

Optional references Lecturer Håkan Hjalmarsson

Signals Stationary stochastic processes. Quasi-stationarity. Convergence properties

Reading material Ljung: Chap 2,3

Optional references Lecturer Håkan Hjalmarsson

Signals Stationary stochastic processes. Quasi-stationarity. Convergence properties

Reading material Ljung: Chap 2

Optional referencesmaterial -¶

Lecturer Håkan Hjalmarsson

Signals Stationary stochastic processes. Quasi-stati Estimation Methods
* Minimum mean-square error estimation
* Prediction, filtering and smoothing.
* Optimal  filtering.
* Orthogonarlity. Convergence propertie condition.
* Wiener/Kalman/particle filters

Reading material Ljung: Chap 2

Optional material -

Lecturer Håkan Hjalmarsson

Signal Estimation Methods
* Minimum mean-square error estimation
* Prediction, filtering and smoothing.
* Optimal  filtering.
* Orthogonality condition.
* Wiener/Kalman/particle filters
Reading material Ljung: Chap 23

Optional material -

Lecturer Håkan Hjalmarsson

Signal Estimation Methods
* Minimum mean-square error estimation
* Prediction, filtering and smoothing.
* Optimal  filtering.
* Orthogonality condition.
* Wiener/Kalman/particle filters
Reading material Ljung: Chapter 3

Optional material -

Lecturer Håkan Hjalmarsson

Signal Estimation Methods
* Minimum mean-square error estimation
* Prediction, filtering and smoothing.
* Optimal  filtering.
* Wiener/Kalman/particle filters
Reading material Ljung: Chapter 3

Optional material -Lemma 2.1

Lecturer Håkan Hjalmarsson

Signal Estimation Methods
* Minimum mean-square error estimation
* Prediction, filtering and smoothing.
* Optimal  filtering.
* Wiener/Kalman/particle filters
Reading material Ljung: Chapter 3

Optional material Lemma 2.1¶ Lecturer Håkan Hjalmarsson

Q11

Introduction Presentation of basic ideas and concepts, review of probability and statistics.

Reading material Ljung: Chap 1, Apprndix. I, II

Optional references  Ljung,2010

 Lecturer Håkan Hjalmarsson

Introduction Presentation of basic ideas and concepts, review of probability . Linear time-invariandt statisticystems.

Reading material Ljung: Chap 1, Apprndix. I, IIters 1, 2.1, 2.2

Optional references  L. Ljung,. “Perspectives on    System Identification”. Annual    Reviews in Control, Vol. 34(1), pp. 1-12, 2010.

 Lecturer Håkan Hjalmarsson

Introduction Presentation of basic ideas and concepts. Linear time-invariant systems.

Reading material Ljung: Chapters 1, 2.1, 2.2

Optional referencesmaterial  L. Ljung. “Perspectives on    System Identification”. Annual    Reviews in Control, Vol. 34(1), pp. 1-12, 2010.

 Lecturer Håkan Hjalmarsson

Introduction Presentation of basic ideas and concepts. Linear time-invariant systems.

Reading material Ljung: Chapters 1, 2.1, 2.2

Optional material  L. Ljung. “Perspectives on    System Identification”. Annual    Reviews in Control, Vol. 34(1), pp. 1-12, 2010.

 Lecturer Håkan Hjalmarsson

Introduction Presentation of basic ideas and concepts. Linear time-invariant systems.

Reading material Ljung: Chapters 1, 2.1, 2.2

Optional material  L. Ljung. “Perspectives on System Identification”. Annual    Reviews in Control, Vol. 34(1), pp. 1-12, 2010.

 Lecturer Håkan Hjalmarsson

Introduction Presentation of b
* Basic ideas and concepts. Linear time-invariant systems.¶
* Models of dynamical systems
* Noise models
* Stochastic processes
Reading material Ljung: Chapters 1, 2.1, 2.2

Optional material  L. Ljung. “Perspectives on System Identification”. Annual  Reviews in Control, Vol. 34(1), pp. 1-12, 2010.

 Lecturer Håkan Hjalmarsson

Introduction
* Basic ideas and concepts.
* Models of dynamical systems
* Noise models
* Stochastic processes
Reading material Ljung: Chapters 1, 2.1, 2.2

Optional material  L. Ljung. “Perspectives on System Identification”. Annual  Reviews in Control, Vol. 34(1), pp. 1-12, 2010.

 Lecturer Håkan Hjalmarsson

Q11

Lecture 143

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Closed loop identification Ljung: Chapter 13.4-13.5¶

Reading Material Ljung: Chapter 16

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Closed loop identification Ljung: Chapter 8.5, 13.4-13.5

Reading Material Ljung: Chapter 16

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Closed loop identification Ljung: Chapter 8.5, 13.4-13.5

Reading Material Ljung: Chapter 16

Optional References
* Slides 10
* Slides 11
Lecturer Cristian RojasLecturer Håkan Hjalmarsson

Lecture 132

L21

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Model Selection and Model VaOpen Loop Experiment Design
* Basics

* Informative experiments
* Choice of sampling interval and pre-sampling filters
* Applidcation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16s oriented experiment design 

* Relating application specifications to the identification specifications.
* Least-costly identification
* Computational issue

* Convexification
* Frequency-domain design
*  The cost of system and model complexity.
* Robust design
* Adaptive
Reading Material Ljung: Chapter 13.1-13.3, Applications oriented experiment design

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Open Loop Experiment Design
* Basics

* Informative experiments
* Choice of sampling interval and pre-sampling filters
* Applications oriented experiment design 

* Relating application specifications to the identification specifications.
* Least-costly identification
* Computational issue

* Convexification
* Frequency-domain design
*  The cost of system and model complexity.
* Robust design
* Adaptive
Reading Material Ljung: Chapter 13.1-13.3, Applications oriented experiment design

Optional References
* Slides 10
* Slides 11
Lecturer Cristian RojasLecturer Håkan Hjalmarsson

Lecture 121

L43

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Model Selection and Model ValidEstimation Algorithms
* Solving normal equations using QR-factorization
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis test Non-linear optimization
* Two- and multi-stage methods
Reading Material Ljung: Chapter 160

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Estimation Algorithms
* Solving normal equations using QR-factorization
*  Non-linear optimization
* Two- and multi-stage methods
* The EM-algorithm
Reading Material Ljung: Chapter 10

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Estimation Algorithms
* Solving normal equations using QR-factorization
*  Non-linear optimization
* Two- and multi-stage methods
* The EM-algorithm
Reading Material Ljung: Chapter 10

Optional References
* Slides 10
* Slides 11
Lecturer Cristian RojasHåkan Hjalmarsson

Estimation Algorithms
* Solving normal equations using QR-factorization
*  Non-linear optimization
* Two- and multi-stage methods
* The EM-algorithm
Reading Material Ljung: Chapter 10

Optional References
* Slides 10
* Slides 11
Lecturer Håkan Hjalmarsson

Estimation Algorithms
* Solving normal equations using QR-factorization
*  Non-linear optimization
* Two- and multi-stage methods
* The EM-algorithm
Reading Material Ljung: Chapter 10

Optional References Lecturer Håkan HjalmarssonCristian Rojas

Q11

Lecture 110

Estimation Algorithms
* Solving normal equations using QR-factorization
*  Non-linear optimization
* Two- and multi-stage methods
* The EM-algorithm
Reading Material Ljung: Chapter 10

Optional ReferencesLecture Slides
* Lecture 10¶


Lecturer Cristian Rojas

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Model Selection and Model ValidaIdentification of non-linear systems
* Linear regressions
* Function expansions

* Radial basis function s
* Classic model order selection  criteria (AIC, BIC, MDL),Wavelet expansions
* Kernel expansions
* Splines
*  Hypothesis testsNon-linear filtering and prediction
Reading Material Ljung: Chapter 165

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Identification of non-linear systems
* Linear regressions
* Function expansions

* Radial basis functions
* Wavelet expansions
* Kernel expansions
* Splines
* Non-linear filtering and prediction
Reading Material Ljung: Chapter 5

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas ¶

Lecturer Håkan Hjalmarsson

Q22

Lecture 109

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16¶

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas¶

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16

Optional References
* Slides 10
* Slides 11
Lecturer Cristian RojasHåkan Hjalmarsson

D33

Model Selection and Model Validation
* Classic model order selection  criteria (AIC, BIC, MDL),
*  Hypothesis tests
Reading Material Ljung: Chapter 16

Optional References
* Slides 10  (old)
* Slides 11 (old)
Lecturer Håkan Hjalmarsson

Lecture 98

Model Structure Selection and ValidationQuality Evaluation
* Variance error quantification
* Confidence intervals
* Bias/variance trade-off
Reading Material Ljung: Chapter 169

Optional References
* Slides 10
* Slides 11
Lecturer Cristian Rojas

Model Quality Evaluation
* Variance error quantification
* Confidence intervals
* Bias/variance trade-off
Reading Material Ljung: Chapter 9

Optional References
* Slides 10
* Slides 11      7
Lecturer Cristian Rojas

L44

Model Quality Evaluation
* Variance error quantification
* Confidence intervals
* Bias/variance trade-off
Reading Material Ljung: Chapter 9

Optional References
* Slides      7Lecture Slides
*
Lecturer Cristian Rojas

Lecture 87

Model Quality Evaluation
* Variance error quantification
* Confidence intervals
* Bias/variance trade-offBias / variance trade-off
* Stein's paradox and biased estimators
* Confidence intervals / regions
* Variance error quantification
* Geometric approach to variance analysis
Reading Material Ljung: Chapter 9

Lecture Slides
*
Lecturer Cristian Rojas

Model Quality Evaluation
* Bias / variance trade-off
* Stein's paradox and biased estimators
* Confidence intervals / regions
* Variance error quantification
* Geometric approach to variance analysis
Reading Material
* Ljung: Chapter 9

* A Geometric Approach to Variance Analysis in System Identification, H. Hjalmarsson and J. Mårtensson, IEEE Transactions on Automatic Control, 56(5), pp. 983-997, 2011.
Lecture Slides
* Lecture 7
Lecturer Cristian Rojas

Data Preprocessing and Choice of Identification CriterionFundamental properties of estimation algorithms
* Identifiability
* Persistence of excitation
* Consistency
Reading Material Ljung: Chapters 14, 154.5-4.6, 8

Optional References Lecturer Cristian Rojas

Fundamental properties of estimation algorithms
* Identifiability
* Persistence of excitation
* Consistency
Reading Material Ljung: Chapters 4.5-4.6, 8

Optional References
* Slides      6
Lecturer Cristian Rojas

Q13

Fundamental properties of estimation algorithms
* Identifiability
* Persistence of excitation
* Consistency
Reading Material Ljung: Chapters 4.5-4.6, 8

Optional References
* Slides      6
Lecturer Cristian Rojas

Lecture 76

Fundamental properties of estimation algorithms
* Identifiability
* Persistence of excitation
* Consistency
* Asymptotic distribution of PEM
Reading Material Ljung: Chapters 4.5-4.6, 8

Optional References
* Lecture 6
Lecturer Cristian Rojas