Chaı̂nes de Markov TripletTriplet Markov Chains

The Hidden Markov Chains (HMC) are widely applied in various problems. This succes is mainly due to the fact that the hidden process can be recovered even in the case of very large set of data. These models have been recetly generalized to ‘Pairwise Markov Chains’ (PMC) model, which admit the same processing power and a better modeling one. The aim of this note is to propose further generalization called Triplet Markov Chains (TMC), in which the distribution of the couple (hidden process, observed process) is the marginal distribution of a Markov chain. Similarly to HMC, we show that posterior marginals are still calculable in Triplets Markov Chains. We provide a necessary and sufficient condition that a TMC is a PMC, which shows that the new model is strictly more general. Furthermore, a link with the Dempster–Shafer fusion is specified. To cite this article: W. Pieczynski, C. R. Acad. Sci. Paris, Ser. I 335 (2002) 275–278.     

Arbres de Markov Triplet et fusion de Dempster–Shafer

The hidden Markov chains (HMC) (X,Y) have been recently generalized to triplet Markov chains (TMC), which enjoy the same capabilities of restoring a hidden process X from the observed process Y. The posterior distribution of X can be viewed, in an HMC, as a particular case of the so called “Dempster–Shafer fusion” (DS fusion) of the prior Markov with a probability q defined from the observation Y=y. As such, when we place ourselves in the Dempster–Shafer theory of evidence by replacing the probability distribution of X by a mass function M having an analogous Markov form (which gives again the classical Markov probability distribution in a particular case), the result of DS fusion of M with q generalizes the conventional posterior distribution of X. Although this result is not necessarily a Markov distribution, it has been recently shown that it is a TMC, which renders traditional restoration methods applicable. The aim of this Note is to present some generalizations of the latter result: (i) more general HMCs can be considered; (ii) q, which can possibly be a mass function Q, is itself a result of the DS fusion; and (iii) all these results are finally specified in the hidden Markov trees (HMT) context, which generalizes the HMC one. To cite this article: W. Pieczynski, C. R. Acad. Sci. Paris, Ser. I 336 (2003).     

Fusion de Dempster–Shafer dans les chaînes triplet partiellement de Markov

Hidden Markov Chains (HMC), Pairwise Markov Chains (PMC), and Triplet Markov Chains (TMC), allow one to estimate a hidden process X from an observed process Y. More recently, TMC have been generalized to Triplet Partially Markov chain (TPMC), where the estimation of X from Y remains workable. Otherwise, when introducing a Dempster–Shafer mass function instead of prior Markov distribution in classical HMC, the result of its Dempster–Shafer fusion with a distribution provided $Y=y$, which generalizes the posterior distribution of X, is a TMC. The aim of this Note is to generalize the latter result replacing HMC with multisensor TPMC. To cite this article: W. Pieczynski, C. R. Acad. Sci. Paris, Ser. I 339 (2004).     

Interpretation of the Evolution of the Homogeneous Markov System (or,equivalently, of the Embedded Markov Chain) as the Deformation of a Viscoelastic Medium. The 3-D Case

Every attainable structure of a continuous time homogeneous Markov chain (HMC) with n states, or of a closed Markov system with an embedded HMC with n states, or more generally of a Markov system driven by an HMC, is considered as a point-particle of ? ⁿ . Then, the motion of the attainable structure corresponds to the motion of the respective point-particle in ? ⁿ . Under the assumption that “the motion of every particle at every time point is due to the interaction with its surroundings”, ? ⁿ (and equivalently the set of the accosiated attainable structures of the homogeneous Markov system (HMS), or alternatively of the underlying embedded HMC) becomes a continuum. Thus, the evolution of the set of the attainable structures corresponds to the motion of the continuum. In this paper it is shown that the evolution of a three-dimensional HMS (n = 3) or simply of an HMC, can be interpreted through the evolution of a two-dimensional isotropic viscoelastic medium.     

Copules gaussiennes dans les chaînes triplet partiellement de Markov

Hidden Markov chains, which are widely used in different data restoration problems, have recently been generalised to pairwise partially Markov chains, in which the hidden chain is no longer necessarily Markovian and the distribution of the observed chain, conditional on the hidden one, is of any form. First, we show the applicability of the models in the Gaussian case, with a particular attention to long range correlation noises. Second, we show that the use of copulas allows one to take into account any other form of marginal distributions of the observed chain, conditionally to the hidden one. We end by extending the latter model to a triplet partially Markov chain case. To cite this article: W. Pieczynski, C. R. Acad. Sci. Paris, Ser. I 341 (2005).     

An S-Plus Implementation of Hidden Markov Models in Continuous Time

Abstract

Hidden Markov models (HMM) can be applied to the study of time varying unobserved categorical variables for which only indirect measurements are available. An S-Plus module to fit HMMs in continuous time to this type of longitudinal data is presented. Covariates affecting the transition intensities of the hidden Markov process or the conditional distribution of the measured response (given the hidden states of the process) are handled under a generalized regression framework. Users can provide C subroutines specifying the parameterization of the model to adapt the software to a wide variety of data types. HMM analysis using the S-Plus module is illustrated on a dataset from a prospective study of human papillomavirus infection in young women and on simulated data.     

Estimation fonctionnelle de la densité de l'opérateur de transition d'un processus de Markov à temps discretFunctional estimate of Markov transition operator density: Discrete time case

In this study, we build a kernel estimate of the transition operator density for some discrete time continuous states Markov processes which satisfy some general conditions. Assumptions are not restrictive and results can be used on transition Markov operator, viewed as an endomorphism on L^p, p∈[1,∞[, as well as on its adjoint. The main result deals with convergence rate of built kernel estimate. To cite this article: A. Laksaci, A. Yousfate, C. R. Acad. Sci. Paris, Ser. I 334 (2002) 1035–1038.     

Phylogenetic Inference for Binary Data on Dendograms Using Markov Chain Monte Carlo

Abstract

Using a stochastic model for the evolution of discrete characters among a group of organisms, we derive a Markov chain that simulates a Bayesian posterior distribution on the space of dendograms. A transformation of the tree into a canonical cophenetic matrix form, with distinct entries along its superdiagonal, suggests a simple proposal distribution for selecting candidate trees “close” to the current tree in the chain. We apply the consequent Metropolis algorithm to published restriction site data on nine species of plants. The Markov chain mixes well from random starting trees, generating reproducible estimates and confidence sets for the path of evolution.     

Markov random fields,Markov cocycles and the 3-colored chessboard

The well-known Hammersley–Clifford Theorem states (under certain conditions) that any Markov random field is a Gibbs state for a nearest neighbor interaction. In this paper we study Markov random fields for which the proof of the Hammersley–Clifford Theorem does not apply. Following Petersen and Schmidt we utilize the formalism of cocycles for the homoclinic equivalence relation and introduce “Markov cocycles”, reparametrizations of Markov specifications. The main part of this paper exploits this to deduce the conclusion of the Hammersley–Clifford Theorem for a family of Markov random fields which are outside the theorem’s purview where the underlying graph is Z_d. This family includes all Markov random fields whose support is the d-dimensional “3-colored chessboard”. On the other extreme, we construct a family of shift-invariant Markov random fields which are not given by any finite range shift-invariant interaction.     

The age of a Markov process

The concept of a limiting conditional age distribution of a continuous time Markov process whose state space is the set of non-negative integers and for which {0} is absorbing is defined as the weak limit as t→∞ of the last time before t an associated “return” Markov process exited from {0} conditional on the state, j, of this process at t. It is shown that this limit exists and is non-defective if the return process is ρ-recurrent and satisfies the strong ratio limit property. As a preliminary to the proof of the main results some general results are established on the representation of the ρ-invariant measure and function of a Markov process. The conditions of the main results are shown to be satisfied by the return process constructed from a Markov branching process and by birth and death processes. Finally, a number of limit theorems for the limiting age as j→∞ are given.     

Contraction mappings underlying undiscounted Markov decision problems

This paper is concerned with the properties of the value-iteration operator0 which arises in undiscounted Markov decision problems. We give both necessary and sufficient conditions for this operator to reduce to a contraction operator, in which case it is easy to show that the value-iteration method exhibits a uniform geometric convergence rate. As necessary conditions we obtain a number of important characterizations of the chain and periodicity structures of the problem, and as sufficient conditions, we give a general “scrambling-type” recurrency condition, which encompasses a number of important special cases. Next, we show that a data transformation turns every unichained undiscounted Markov Renewal Program into an equivalent undiscounted Markov decision problem, in which the value-iteration operator is contracting, because it satisfies this “scrambling-type” condition. We exploit this contraction property in order to obtain lower and upper bounds as well as variational characterizations for the fixed point of the optimality equation and a test for eliminating suboptimal actions.     

On the Conservativeness of Some Markov Processes

We give a unified method to obtain the conservativeness of a class of Markov processes associated with lower bounded semi-Dirichlet forms on L ²(X;m), including symmetric diffusion processes, some non-symmetric diffusion processes and jump type Markov processes on X, where X is a locally compact separable metric space and m is a positive Radon measure on X with full topological support. Using the method, we give an example in each section, providing the conservativeness of the processes, that are given by the “increasingness of the volume of some sets(balls)” and “that of the coefficients on the sets” of the Markov processes.     

可列非齐次隐Markov模型的强大数定律

隐马尔科夫模型被广泛的应用于弱相依随机变量的建模,是研究神经生理学、发音过程和生物遗传等问题的有力工具。研究了可列非齐次隐 Markov 模型的若干性质,得到了这类模型的强大数定律,推广了有限非齐次马氏链的一类强大数定律。     

Construction inconditionnelle de groupes de Galois motiviquesAn unconditional construction of motivic Galois groups

We attach to any “classical” Weil cohomology theory over a field a motivic Galois group, defined up to an inner automorphism. We also study the specialisation of numerical motives and the behaviour of motivic Galois group by specialisation. To cite this article: Y. André, B. Kahn, C. R. Acad. Sci. Paris, Ser. I 334 (2002) 989–994.     

Creditworthiness dynamics and Hidden Markov Models

A dynamic monitoring of credit risky portfolios is described. In the first section, it is shown how a Markov dependence can be used in modelling the borrower's behaviour: a chain of transition probabilities matrices is built in which the states of the dynamic stochastic system are the number of instalments in arrears. In the second part, such a model is generalized in the framework of the Hidden Markov Models to explain how the credit market conditions could affect the borrower's payment process. Numerical examples complete the note.     

Separable solutions for Markov processes in random environments

In this paper we address the problem of efficiently deriving the steady-state distribution for a continuous time Markov chain (CTMC) S evolving in a random environment E. The process underlying E is also a CTMC. S is called Markov modulated process. Markov modulated processes have been widely studied in literature since they are applicable when an environment influences the behaviour of a system. For instance, this is the case of a wireless link, whose quality may depend on the state of some random factors such as the intensity of the noise in the environment. In this paper we study the class of Markov modulated processes which exhibits separable, product-form stationary distribution. We show that several models that have been proposed in literature can be studied applying the Extended Reversed Compound Agent Theorem (ERCAT), and also new product-forms are derived. We also address the problem of the necessity of ERCAT for product-forms and show a meaningful example of product-form not derivable via ERCAT.     

Bayesian Monte Carlo estimation for profile hidden Markov models

Hidden Markov models are used as tools for pattern recognition in a number of areas, ranging from speech processing to biological sequence analysis. Profile hidden Markov models represent a class of so-called “left–right” models that have an architecture that is specifically relevant to classification of proteins into structural families based on their amino acid sequences. Standard learning methods for such models employ a variety of heuristics applied to the expectation-maximization implementation of the maximum likelihood estimation procedure in order to find the global maximum of the likelihood function. Here, we compare maximum likelihood estimation to fully Bayesian estimation of parameters for profile hidden Markov models with a small number of parameters. We find that, relative to maximum likelihood methods, Bayesian methods assign higher scores to data sequences that are distantly related to the pattern consensus, show better performance in classifying these sequences correctly, and continue to perform robustly with regard to misspecification of the number of model parameters. Though our study is limited in scope, we expect our results to remain relevant for models with a large number of parameters and other types of left–right hidden Markov models.     

MEXIT: Maximal un-coupling times for stochastic processes

Classical coupling constructions arrange for copies of the same Markov process started at two different initial states to become equal as soon as possible. In this paper, we consider an alternative coupling framework in which one seeks to arrange for two different Markov (or other stochastic) processes to remain equal for as long as possible, when started in the same state. We refer to this “un-coupling” or “maximal agreement” construction as MEXIT, standing for “maximal exit”. After highlighting the importance of un-coupling arguments in a few key statistical and probabilistic settings, we develop an explicit MEXIT construction for stochastic processes in discrete time with countable state-space. This construction is generalized to random processes on general state-space running in continuous time, and then exemplified by discussion of MEXIT for Brownian motions with two different constant drifts.     

Randomizing properties of convex high-dimensional bodies and some geometric inequalitiesLes propriétés aléatoires des corps convexes de grande dimension et une inégalité géométrique

Properties of convex bodies related to uniform distribution are studied. In particular, a low bound for the norm of the sum of independent geometrically distributed vectors is obtained. It extends the previously studied case of identically distributed vectors by Bourgain, Meyer, Milman and Pajor and solves a problem raised there. Another corollary asserts that any finite dimensional normed space has a “random cotype 2”. To cite this article: E. Gluskin, V. Milman, C. R. Acad. Sci. Paris, Ser. I 334 (2002) 875–879.