Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Learning Stochastic Nonlinear Dynamical Systems Using Non-stationary Linear Predictors
KTH, Skolan för elektro- och systemteknik (EES), Reglerteknik. (System Identification)ORCID-id: 0000-0001-5474-7060
2017 (Engelska)Licentiatavhandling, monografi (Övrigt vetenskapligt)
Abstract [en]

The estimation problem of stochastic nonlinear parametric models is recognized to be very challenging due to the intractability of the likelihood function. Recently, several methods have been developed to approximate the maximum likelihood estimator and the optimal mean-square error predictor using Monte Carlo methods. Albeit asymptotically optimal, these methods come with several computational challenges and fundamental limitations.

The contributions of this thesis can be divided into two main parts. In the first part, approximate solutions to the maximum likelihood problem are explored. Both analytical and numerical approaches, based on the expectation-maximization algorithm and the quasi-Newton algorithm, are considered. While analytic approximations are difficult to analyze, asymptotic guarantees can be established for methods based on Monte Carlo approximations. Yet, Monte Carlo methods come with their own computational difficulties; sampling in high-dimensional spaces requires an efficient proposal distribution to reduce the number of required samples to a reasonable value.

In the second part, relatively simple prediction error method estimators are proposed. They are based on non-stationary one-step ahead predictors which are linear in the observed outputs, but are nonlinear in the (assumed known) input. These predictors rely only on the first two moments of the model and the computation of the likelihood function is not required. Consequently, the resulting estimators are defined via analytically tractable objective functions in several relevant cases. It is shown that, under mild assumptions, the estimators are consistent and asymptotically normal. In cases where the first two moments are analytically intractable due to the complexity of the model, it is possible to resort to vanilla Monte Carlo approximations. Several numerical examples demonstrate a good performance of the suggested estimators in several cases that are usually considered challenging.

Ort, förlag, år, upplaga, sidor
Stockholm, Sweden: KTH Royal Institute of Technology, 2017. , s. 186
Serie
TRITA-EE, ISSN 1653-5146 ; 2017:172
Nyckelord [en]
Stochastic Nonlinear Systems, Nonlinear System Identification, Learning Dynamical Models, Maximum Likelihood, Estimation, Process Disturbance, Prediction Error Method, Non-stationary Linear Predictors, Intractable Likelihood, Latent Variable Models
Nationell ämneskategori
Reglerteknik Signalbehandling Annan elektroteknik och elektronik Sannolikhetsteori och statistik
Forskningsämne
Elektro- och systemteknik
Identifikatorer
URN: urn:nbn:se:kth:diva-218100ISBN: 978-91-7729-624-9 (tryckt)OAI: oai:DiVA.org:kth-218100DiVA, id: diva2:1160265
Presentation
2017-12-20, B:218 (Hörsal Q2), Osquldas väg 10, Q-huset, våningsplan 2, KTH Campus, Stockholm, 10:00 (Engelska)
Opponent
Handledare
Anmärkning

QC 20171128

Tillgänglig från: 2017-11-28 Skapad: 2017-11-25 Senast uppdaterad: 2017-11-28Bibliografiskt granskad

Open Access i DiVA

fulltext(4929 kB)340 nedladdningar
Filinformation
Filnamn FULLTEXT01.pdfFilstorlek 4929 kBChecksumma SHA-512
9ff22c40c2d3a772a6b170a79f2818c2cffbc574dc658f8548d9611eb709466ad243a3f8e6758adc2c315dd1f00b6d1c1b60b9f6a24cafe9ebd5060c9df4dcd9
Typ fulltextMimetyp application/pdf

Sök vidare i DiVA

Av författaren/redaktören
Abdalmoaty, Mohamed
Av organisationen
Reglerteknik
ReglerteknikSignalbehandlingAnnan elektroteknik och elektronikSannolikhetsteori och statistik

Sök vidare utanför DiVA

GoogleGoogle Scholar
Totalt: 340 nedladdningar
Antalet nedladdningar är summan av nedladdningar för alla fulltexter. Det kan inkludera t.ex tidigare versioner som nu inte längre är tillgängliga.

isbn
urn-nbn

Altmetricpoäng

isbn
urn-nbn
Totalt: 5551 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf