DocumentCode :
828492
Title :
Necessary and sufficient conditions for local second-order identifiability
Author :
Goodrich, R.L. ; Caines, P.E.
Author_Institution :
ABT Associates, Incorporated, Cambridge, MA, USA
Volume :
24
Issue :
1
fYear :
1979
fDate :
2/1/1979 12:00:00 AM
Firstpage :
125
Lastpage :
127
Abstract :
This note presents a new formulation and proof of the result the Hessian of the likelihood function of an observed process at the point 
in a parameter space, computed under the assumption that the process is i.i.d. Gaussian, is asymptotically nonsingular if and only if is locally second-order identifiable. That is to say, if and only if the parameters in a neighborhood of are in one-to-one correspondence with the second-order statistics of the observed process.
in a parameter space, computed under the assumption that the process is i.i.d. Gaussian, is asymptotically nonsingular if and only if is locally second-order identifiable. That is to say, if and only if the parameters in a neighborhood of are in one-to-one correspondence with the second-order statistics of the observed process.Keywords :
Hessian matrices; Parameter identification; maximum-likelihood (ML) estimation; Adaptive control; Convergence; Equations; Filters; Linear systems; Observers; Parameter estimation; Programmable control; Sufficient conditions; System identification;
fLanguage :
English
Journal_Title :
Automatic Control, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9286
Type :
jour
DOI :
10.1109/TAC.1979.1101953
Filename :
1101953
Link To Document :
