论文信息 - On local maxima of the likelihood function for Toeplitz matrix estimation

On local maxima of the likelihood function for Toeplitz matrix estimation

It is shown that in estimating a Toeplitz covariance matrix, the likelihood function can present a local maximum close to the global maximum. An analytical proof is given for the matrices of order 3, and numerical examples are shown for orders 3, 4, and 5. >

Serge Dégerine

[1] T. Barnard,et al. Two maximum entropy beamforming algorithms for equally spaced line arrays , 1982 .

[2] A Nguyen. On the uniqueness of the maximum-likeliwood estimate of structured covariance matrices , 1984 .

[3] Pham Dinh Tuan,et al. Maximum likelihood estimation of the autoregressive model by relaxation on the reflection coefficients , 1988, IEEE Trans. Acoust. Speech Signal Process..

[4] D.T. Pham,et al. Efficient computation of autoregressive estimates through a sufficient statistic , 1990, IEEE Trans. Acoust. Speech Signal Process..

[5] D. Luenberger,et al. Estimation of structured covariance matrices , 1982, Proceedings of the IEEE.

[6] P. Lopez,et al. 01 - Estimation d'une matrice interspectrale de structure imposée. Applications , 1984 .

[7] Serge Degerine. MAXIMUM LIKELIHOOD ESTIMATION OF AUTOCOVARIANCE MATRICES FROM REPLICATED SHORT TIME SERIES , 1987 .