Soit un noyau markovien sur un espace mesurable muni d’une tribu à base dénombrable, soit tel que , avec , et soit l’espace des fonctions mesurables de dans telles que . Nous démontrons que est quasi-compact sur si et seulement si, pour tout , contient une sous-suite convergeant dans vers , où est une fonction mesurable positive bornée sur et une probabilité sur . En particulier, quand le sous-espace de constitué des fonctions -invariantes est de dimension finie, la convergence uniforme des moyennes est équivalente à la convergence ponctuelle.
Let be a Markov kernel on a measurable space with countably generated -algebra, let such that with , and let be the space of measurable functions on satisfying . We prove that is quasi-compact on if and only if, for all , contains a subsequence converging in to , where the ’s are non-negative bounded measurable functions on and the ’s are probability distributions on . In particular, when the space of -invariant functions in is finite-dimensional, uniform ergodicity is equivalent to mean ergodicity.
@article{AIHPB_2008__44_6_1090_0, author = {Herv\'e, Lo\"\i c}, title = {Quasi-compactness and mean ergodicity for Markov kernels acting on weighted supremum normed spaces}, journal = {Annales de l'I.H.P. Probabilit\'es et statistiques}, volume = {44}, year = {2008}, pages = {1090-1095}, doi = {10.1214/07-AIHP145}, mrnumber = {2469336}, zbl = {1186.37014}, language = {en}, url = {http://dml.mathdoc.fr/item/AIHPB_2008__44_6_1090_0} }
Hervé, Loïc. Quasi-compactness and mean ergodicity for Markov kernels acting on weighted supremum normed spaces. Annales de l'I.H.P. Probabilités et statistiques, Tome 44 (2008) pp. 1090-1095. doi : 10.1214/07-AIHP145. http://gdmltest.u-ga.fr/item/AIHPB_2008__44_6_1090_0/
[1] Quelques applications probabilistes de la quasi-compacité. Ann. Inst. H. Poincaré, Sect. B (N.S.) 10 (1974) 301-337. | Numdam | MR 373008 | Zbl 0318.60064
and .[2] Linear Operators. Part. I: General Theory. Wiley, New York, 1958. | MR 1009162 | Zbl 0084.10402
and .[3] Quasi-compactness and absolutely continuous kernels. Probab. Theory Related Fields. 139 (2007) 451-471. | MR 2322704 | Zbl 1128.60061
.[4] Quasi-compactness and absolutely continuous kernels. Applications to Markov chains (2006). Available at ArXiv:math.PR/0606680. | MR 2322704 | Zbl 1128.60061
.[5] On ergodicity and recurrence properties of a Markov chain with an application to an open Jackson network. Adv. in Appl. Probab. 24 (1992) 343-376. | MR 1167263 | Zbl 0766.60085
and .[6] Transition probabilities and contractions of L∞. Z. Wahrsch. Verw. Gebiete 24 (1972) 263-274. | MR 331516 | Zbl 0228.60028
.[7] Ergodic Theorems. de Gruyter Studies in Mathematics, de Gruyter, Berlin, 1985. | MR 797411 | Zbl 0575.28009
.[8] Quasi-compactness and uniform ergodicity of Markov operators. Ann. Inst. H. Poincaré, Sect. B (N.S.) 11 (1975) 345-354. | Numdam | MR 402007 | Zbl 0318.60065
.[9] Quasi-compactness and uniform ergodicity of positive operators. Israel J. Math. 29 (1978) 309-311. | MR 493502 | Zbl 0374.47015
.[10] Markov Chains and Stochastic Stability. Springer, London, 1993. | MR 1287609 | Zbl 0925.60001
and .[11] Markov Chains. North-Holland, Amsterdam, 1975. | MR 758799 | Zbl 0332.60045
.[12] Topological Vector Spaces. Springer, New York, 1971. | MR 342978 | Zbl 0217.16002
.[13] Banach Lattices and Positive Operators. Springer, New York, 1974. | MR 423039 | Zbl 0296.47023
.