Summary
This paper presents a maximum likelihood estimation method for imperfectly observed Gibbsian fields on a finite lattice. This method is an adaptation of the algorithm given in Younes [28]. Presentation of the new algorithm is followed by a theorem about the limit of the second derivative of the likelihood when the lattice increases, which is related to convergence of the method. Some practical remarks about the implementation of the procedure are eventually given.
Article PDF
Similar content being viewed by others
References
Benviste, A., Métivier, M., Priouret, B.: Algorithmes adaptatifs et approximations stochastiques. Techniques stochastiques. Paris: Masson 1987
Besag, J.: Spatial interaction and the statistical analysis of lattice systems. J. Roy. Stat. Soc. B 36, 192–236 (1974)
Besag, J.: On the statistical analysis of dirty pictures. J. Roy. Stat. Soc. B 48, 259–303 (1986)
Chalmond, B.: Image restoration using an estimated Markov model. Preprint Université Paris Sud (1988)
Dobrushin, R.L.: The description of a random field by means of conditional probabilities and conditions of its regularity. Theory Probab. Appl. 8, 197–224 (1968)
Föllmer, H.: A covariance estimate for Gibbs measures. J. Funct. Anal. 46, 387–395 (1982)
Geman, D., Geman, S.: Stochastic relaxation, Gibbs distribution and Bayesian restoration of images. IEEE TPAMI, vol. PAMI 6, pp. 721–741 (1984)
Geman, S., Graffigne, C.: Markov random field image models and their applications to computer vision. Proceedings of the international congress of mathematicians. Gleason, A.M. (ed.). Providence: AMS 1987
Gidas, B.: Paramter estimation for Gibbs distributions. In: Fully observed data. Preprint, Brown University (1988)
Guyon, X.: Parameter estimation for a stationary process on a d-dimensional lattice. Biometrika 69, 95–105 (1982)
Guyon, X.: Pseudo maximum de vraisemblance et champs Markoviens. In: Droesbeke, F. (ed.) Spatial processes and spatial time series analysis. Proc. 6th. Franco-Belgian Meeting of Statisticians. Publication des Pascaltés Universitaires de Saint Louis-Bruxelles 15–62 (1987)
Guyon, X.: Estimation d'un Champ de Gibbs. Preprint, Université Paris Sud (1986)
Künsch, H.: Thermodynamics and the statistical analysis of Gaussian random fields. Z. Wahrscheinlichkeitstheor. Verw. Geb. 58, 407–421 (1981)
Künsch, H.: Decay of correlations under Dobrushin's uniqueness condition and its applications. Commun. Math. Phys. 84, 207–222 (1982)
Künsch, H.: Asymptotically unbiased inference for Ising models. J. Appl. Probab. 19, 345–357 (1982)
Künsch, H.: Discussion on Besag's paper [3]
Lippmann, A.: A maximum entropy method for expert systems. Brown University Thesis (1986)
Métivier, M., Priouret, P.: Théorèmes de convergence presque sûre pour une classe d'algorithmes stochastiques à pas décroissant. Probab. Th. Rel. Fields 74, 403–428 (1987)
Xanh, N.X., Zessin, H.: Ergodic theorems for spatial processes. Z. Wahrscheinlichkeitstheor. Verw. Geb. 48, 133–158 (1979)
Pickard, D.: Asymptotic inference for an Ising lattice. J. Appl. Probab. 13, 486–497 (1976)
Pickard, D.: Asymptotic inference for an Ising lattice, II. Adv. Appl. Probab. 9, 479–501 (1977)
Pickard, D.: Asymptotic inference for an Ising lattice, III. J. Appl. Probab. 16, 12–24 (1979)
Pickard, D.: Inference for general Ising models. J. Appl. Probab. 19A, 345–357 (1982)
Possolo, A.: Estimation of binary Markov random fields. University of Washington, Technical Report (1986)
Preston, C.: Random fields. (Lect. Notes Math., vol. 534) Berlin Heidelberg New York: Springer 1976
Simon, B.: A remark on Dobrushin uniqueness theorem. Commun. Math. Phys. 183–185 (1979)
Younes, L.: Couplage de l'estimation et du recuit pour des champs de Gibbs. C.R.Acad. Sci. Paris, Ser. I, 303, 659–662 (1986)
Younes, L.: Estimation and annealing for Gibbsian fields. Ann. Inst. Henri Poincaré 24, 269–294 (1988)
Younes, L.: Estimation pour des Champs de Gibbs et application au traitement d'images. Université Paris-Sud, Thesis (1988)
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Younes, L. Parametric Inference for imperfectly observed Gibbsian fields. Probab. Th. Rel. Fields 82, 625–645 (1989). https://doi.org/10.1007/BF00341287
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF00341287