The paper proposes an adjusted maximum likelihood estimator for the parametric estimate of a STARG(p, l0,...,lp) model with measurement noise. Provided the noise variance is known or can be estimated consistently, the adjusted maximum likelihood estimator is proved to be asymptotically equivalent to the corresponding exact maximum likelihood estimator that, in this study, turns out to be computationally untractable. The theoretic background outlined in the paper finds a natural field of application in observed image sequences. Thus, we present the results of a state-space smoothing procedure performed on monthly observations over a regular lattice.
Space time Noisy Observation Smoothing
ROMAGNOLI, Luca
2004-01-01
Abstract
The paper proposes an adjusted maximum likelihood estimator for the parametric estimate of a STARG(p, l0,...,lp) model with measurement noise. Provided the noise variance is known or can be estimated consistently, the adjusted maximum likelihood estimator is proved to be asymptotically equivalent to the corresponding exact maximum likelihood estimator that, in this study, turns out to be computationally untractable. The theoretic background outlined in the paper finds a natural field of application in observed image sequences. Thus, we present the results of a state-space smoothing procedure performed on monthly observations over a regular lattice.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
