Processes with long memory: Regenerative construction and perfect simulation

We present a perfect simulation algorithm for stationary processes indexed by Z, with summable memory decay. Depending on the decay, we construct the process on finite or semiinfinite intervals, explicitly from an i.i.d. uniform sequence. Even though the process has infinite memory, its value at time 0 depends only on a finite, but random, number of these uniform variables. The algorithm is based on a recent regenerative construction of these measures by Ferrari, Maass, Mart´onez and Ney. As applications, we discuss the perfect simulation of binary autoregressions and Markov chains on the unit interval.

[1]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[2]  D. Ornstein.  Ergodic theory, randomness, and dynamical systems , 1974 .

[3]  Benjamin Weiss,et al.  How Sampling Reveals a Process , 1990 .

[4]  Xavier Guyon,et al.  Random fields on a network , 1995 .

[5]  M. Barnsley,et al.  Invariant measures for Markov processes arising from iterated function systems with place-dependent , 1988 .

[6]  Radu Theodorescu,et al.  General control systems , 1978, Inf. Sci..

[7]  E. Nummelin,et al.  A splitting technique for Harris recurrent Markov chains , 1978 .

[8]  M. Frank Norman,et al.  An ergodic theorem for evolution in a random environment , 1975, Journal of Applied Probability.

[9]  James Allen Fill,et al.  An interruptible algorithm for perfect sampling via Markov chains , 1997, STOC '97.

[10]  Alejandro Maass,et al.  Cesàro mean distribution of group automata starting from measures with summable decay , 2000, Ergodic Theory and Dynamical Systems.

[11]  Thomas Kaijser.  A Limit Theorem for Partially Observed Markov Chains , 1975 .

[12]  Peter Walters,et al.  Ruelle’s operator theorem and $g$-measures , 1975 .

[13]  David Bruce Wilson,et al.  Annotated bibliography of perfectly random sampling with Markov chains , 1997, Microsurveys in Discrete Probability.

[14]  Decay of Correlations for Non-Hölderian Dynamics. A Coupling Approach , 1998, math/9806132.

[15]  A. Galves,et al.  Speed of d-convergence for Markov approximations of chains with complete connections. A coupling approach☆ , 1999 .

[16]  J. Propp,et al.  Exact sampling with coupled Markov chains and applications to statistical mechanics , 1996 .

[17]  Regeneration for chains with infinite memory , 1993 .

[18]  Recent Advances in the Metric Theory of Continued Fractions , 1978 .

[19]  R. Tweedie,et al.  Perfect simulation and backward coupling , 1998 .

[20]  W. Doeblin.  Remarques sur la théorie métrique des fractions continues , 1940 .

[21]  H. Berbee,et al.  Chains with infinite connections: Uniqueness and Markov representation , 1987 .

[22]  David Bruce Wilson,et al.  Exact sampling with coupled Markov chains and applications to statistical mechanics , 1996, Random Struct. Algorithms.

[23]  W. Doeblin.  Sur deux problèmes de M. Kolmogoroff concernant les chaînes dénombrables , 1938 .

[24]  M. Norman.  Markovian Learning Processes , 1974 .

[25]  Robert Fortet,et al.  Sur des chaînes à liaisons complètes , 1937 .

[26]  Marius Iosifescu,et al.  Dependence with Complete Connections and its Applications , 1990 .

[27]  Editors , 1986, Brain Research Bulletin.

[28]  J. A. Fill.  An interruptible algorithm for perfect sampling via Markov chains , 1998 .

[29]  J. N. Corcoran,et al.  Perfect Sampling of Harris Recurrent Markov Chains , 1999 .

[30]  N E Manos,et al.  Stochastic Models , 1960, Encyclopedia of Social Network Analysis and Mining. 2nd Ed..

[31]  P. McCullagh,et al.  Generalized Linear Models , 1984 .

[32]  R. Tweedie,et al.  Perfect sampling of ergodic Harris chains , 2001 .

[33]  B. M. Fulk.  MATH , 1992 .

[34]  C. Caramanis.  What is ergodic theory , 1963 .

[35]  Radu Theodorescu,et al.  Random processes and learning , 1969 .

[36]  Thomas Kaijser.  On a theorem of Karlin , 1994 .

[37]  F. Ledrappier.  Principe variationnel et systèmes dynamiques symboliques , 1974 .

[38]  S. Lalley,et al.  Regeneration in One-Dimensional Gibbs States and Chains with Complete Connections , 2000 .

[39]  K. Athreya,et al.  A New Approach to the Limit Theory of Recurrent Markov Chains , 1978 .

[40]  Verzekeren Naar Sparen,et al.  Cambridge , 1969, Humphrey Burton: In My Own Time.

[41]  John A. Nelder,et al.  Generalized linear models. 2nd ed. , 1993 .

[42]  Steven P. Lalley,et al.  Regenerative Representation for One-Dimensional Gibbs States , 1986 .

[43]  T. E. Harris,et al.  On chains of infinite orde , 1955 .

[44]  Andrew G. Glen,et al.  APPL , 2001 .

[45]  Functions of event variables of a random system with complete connections , 1977 .

[46]  Iterated function systems a rising from recursive estimation problems , 1992 .