A New Method for Evaluating the Log-Likelihood Gradient (Score) of Linear Dynamic Systems

Mordechai Segal; Ehud Weinstein

doi:10.1109/9.1295

A New Method for Evaluating the Log-Likelihood Gradient (Score) of Linear Dynamic Systems

Mordechai Segal, Ehud Weinstein

School of Electrical Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

A new method for evaluating the log-likelihood gradient (score) of linear dynamic systems is presented, based on the optimal smoothing equations. The new result can be used for efficient calculations and approximations of gradient-search algorithms for maximum likelihood estimation of the unknown system parameters.

Original language	English
Pages (from-to)	763-766
Number of pages	4
Journal	IEEE Transactions on Automatic Control
Volume	33
Issue number	8
DOIs	https://doi.org/10.1109/9.1295
State	Published - Aug 1988

Access to Document

10.1109/9.1295

Cite this

@article{08118087cb7b4fb59d808316344a115a,

title = "A New Method for Evaluating the Log-Likelihood Gradient (Score) of Linear Dynamic Systems",

abstract = "A new method for evaluating the log-likelihood gradient (score) of linear dynamic systems is presented, based on the optimal smoothing equations. The new result can be used for efficient calculations and approximations of gradient-search algorithms for maximum likelihood estimation of the unknown system parameters.",

author = "Mordechai Segal and Ehud Weinstein",

year = "1988",

month = aug,

doi = "10.1109/9.1295",

language = "אנגלית",

volume = "33",

pages = "763--766",

journal = "IEEE Transactions on Automatic Control",

issn = "0018-9286",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "8",

}

TY - JOUR

T1 - A New Method for Evaluating the Log-Likelihood Gradient (Score) of Linear Dynamic Systems

AU - Segal, Mordechai

AU - Weinstein, Ehud

PY - 1988/8

Y1 - 1988/8

N2 - A new method for evaluating the log-likelihood gradient (score) of linear dynamic systems is presented, based on the optimal smoothing equations. The new result can be used for efficient calculations and approximations of gradient-search algorithms for maximum likelihood estimation of the unknown system parameters.

AB - A new method for evaluating the log-likelihood gradient (score) of linear dynamic systems is presented, based on the optimal smoothing equations. The new result can be used for efficient calculations and approximations of gradient-search algorithms for maximum likelihood estimation of the unknown system parameters.

UR - http://www.scopus.com/inward/record.url?scp=0024054707&partnerID=8YFLogxK

U2 - 10.1109/9.1295

DO - 10.1109/9.1295

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0024054707

SN - 0018-9286

VL - 33

SP - 763

EP - 766

JO - IEEE Transactions on Automatic Control

JF - IEEE Transactions on Automatic Control

IS - 8

ER -

A New Method for Evaluating the Log-Likelihood Gradient (Score) of Linear Dynamic Systems

Abstract

Access to Document

Other files and links

Fingerprint

Cite this