Geodesic
Marginal problem, statistical estimation, and Möbius formula
Kybernetika, Tome 43 (2007) no. 5, pp. 619-631 Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

A solution to the marginal problem is obtained in a form of parametric exponential (Gibbs–Markov) distribution, where the unknown parameters are obtained by an optimization procedure that agrees with the maximum likelihood (ML) estimate. With respect to a difficult performance of the method we propose also an alternative approach, providing the original basis of marginals can be appropriately extended. Then the (numerically feasible) solution can be obtained either by the maximum pseudo-likelihood (MPL) estimate, or directly by Möbius formula.
A solution to the marginal problem is obtained in a form of parametric exponential (Gibbs–Markov) distribution, where the unknown parameters are obtained by an optimization procedure that agrees with the maximum likelihood (ML) estimate. With respect to a difficult performance of the method we propose also an alternative approach, providing the original basis of marginals can be appropriately extended. Then the (numerically feasible) solution can be obtained either by the maximum pseudo-likelihood (MPL) estimate, or directly by Möbius formula.
Classification : 60G60, 62F10, 62G07, 62H12, 93E12, 93E25
Keywords: Gibbs distributions; maximum entropy; pseudo-likelihood; Möbius formula 
@article{KYB_2007_43_5_a2,
     author = {Jan\v{z}ura, Martin},
     title = {Marginal problem, statistical estimation, and {M\"obius} formula},
     journal = {Kybernetika},
     pages = {619--631},
     year = {2007},
     volume = {43},
     number = {5},
     mrnumber = {2376327},
     zbl = {1138.93059},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/KYB_2007_43_5_a2/}
}
TY  - JOUR
AU  - Janžura, Martin
TI  - Marginal problem, statistical estimation, and Möbius formula
JO  - Kybernetika
PY  - 2007
SP  - 619
EP  - 631
VL  - 43
IS  - 5
UR  - http://geodesic.mathdoc.fr/item/KYB_2007_43_5_a2/
LA  - en
ID  - KYB_2007_43_5_a2
ER  - 
%0 Journal Article
%A Janžura, Martin
%T Marginal problem, statistical estimation, and Möbius formula
%J Kybernetika
%D 2007
%P 619-631
%V 43
%N 5
%U http://geodesic.mathdoc.fr/item/KYB_2007_43_5_a2/
%G en
%F KYB_2007_43_5_a2
Janžura, Martin. Marginal problem, statistical estimation, and Möbius formula. Kybernetika, Tome 43 (2007) no. 5, pp. 619-631. http://geodesic.mathdoc.fr/item/KYB_2007_43_5_a2/

[1] Barndorff-Nielsen O. E.: Information and Exponential Families in Statistical Theory. Wiley, New York 1978 | MR | Zbl

[2] Besag J.: Statistical analysis of non-lattice data. The Statistician 24 (1975), 179–195

[3] Csiszár I., Matúš F.: Generalized maximum likelihood estimates for exponential families. Probability Theory and Related Fields (to appear) | MR | Zbl

[4] Dobrushin R. L.: Prescribing a system of random variables by conditional distributions. Theor. Probab. Appl. 15 (1970), 458–486 | Zbl

[5] Gilks W. R., Richardson, S., (eds.) D. J. Spiegelhalter: Markov Chain Monte Carlo in Practice. Chapman and Hall, London 1996 | MR | Zbl

[6] Janžura M.: Asymptotic results in parameter estimation for Gibbs random fields. Kybernetika 33 (1997), 2, 133–159 | MR | Zbl

[7] Janžura M.: A parametric model for large discrete stochastic systems. In: Second European Conference on Highly Structured Stochastic Systems, Pavia 1999, pp. 148–150

[8] Janžura M., Boček P.: A method for knowledge integration. Kybernetika 34 (1988), 1, 41–55

[9] Jaynes E. T.: On the rationale of the maximum entropy methods. Proc. IEEE 70 (1982), 939–952

[10] Jiroušek R., Vejnarová J.: Construction of multidimensional model by operators of composition: Current state of art. Soft Computing 7 (2003), 328–335

[11] Lauritzen S. L.: Graphical Models. University Press, Oxford 1006 | MR | Zbl

[12] Perez A.: $\varepsilon $-admissible simplifications of the dependence structure of random variables. Kybernetika 13 (1979), 439–449 | MR

[13] Perez A., Studený M.: Comparison of two methods for approximation of probability distributions with prescribed marginals. Kybernetika 43 (2007), 5, 591–618 | MR | Zbl

[14] Winkler G.: Image Analysis, Random Fields and Dynamic Monte Carlo Methods. Springer–Verlag, Berlin 1995 | MR | Zbl

[15] Younes L.: Estimation and annealing for Gibbsian fields. Ann. Inst. H. Poincaré 24 (1988), 2, 269–294 | MR | Zbl