论文信息 - Maximum likelihood estimation for Gibbsian fields

Maximum likelihood estimation for Gibbsian fields

This paper presents practical remarks and numerical results on maximum likelihood estimation algorithms for perfectly and imperfectly observed Gibbsian fields on a finite lattice. These remarks are preceded by the definition of the algorithms. In the appendix, consistency of maximum likelihood estimation is proved. Math. Reviews Classification: Primary:62F 10 Secondary:60G80.

L. Younes

[1] P. L. Dobruschin. The Description of a Random Field by Means of Conditional Probabilities and Conditions of Its Regularity , 1968 .

[2] J. Besag. Spatial Interaction and the Statistical Analysis of Lattice Systems , 1974 .

[3] H. Zessin,et al. Ergodic theorems for spatial processes , 1979 .

[4] David K. Pickard. Inference for general Ising models , 1982 .

[5] H. Künsch. Decay of correlations under Dobrushin's uniqueness condition and its applications , 1982 .

[6] H. Künsch. Asymptotically unbiased inference for Ising models , 1983, Advances in Applied Probability.

[7] Donald Geman,et al. Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8] B. Gidas. Nonstationary Markov chains and convergence of the annealing algorithm , 1985 .

[9] J. Besag. On the Statistical Analysis of Dirty Pictures , 1986 .

[10] L. Younes. Estimation and annealing for Gibbsian fields , 1988 .