Strong law of large numbers for generalized sample relative entropy of non homogeneous Markov chains

In this paper, we study the strong law of large numbers for the generalized sample relative entropy of non homogeneous Markov chains taking values from a finite state space. First, we introduce the definitions of generalized sample relative entropy and generalized sample relative entropy rate. Then, using a strong limit theorem for the delayed sums of the functions of two variables and a strong law of large numbers for non homogeneous Markov chains, we obtain the strong law of large numbers for the generalized sample relative entropy of non homogeneous Markov chains. As corollaries, we obtain some important results.     

The strong law of large numbers for non homogeneous M-bifurcating Markov chains indexed by a M-branch Cayley tree

This article is devoted to the strong law of large numbers and the entropy ergodic theorem for non homogeneous M-bifurcating Markov chains indexed by a M-branch Cayley tree, which generalizes the relevant results of tree-indexed nonhomogeneous bifurcating Markov chains. Meanwhile, our proof is quite different from the traditional method.     

The definition of tree-indexed Markov chains in random environment and their existence

This paper gives the definition of tree-indexed Markov chains in random environment with discrete state space, and then studies some equivalent theorems of tree-indexed Markov chains in random environment. Finally, we give the equivalence on tree-indexed Markov chains in Markov environment and double Markov chains indexed by a tree.     

Cumulative ratio information based on general cumulative entropy

In this paper, we suggest an extension of the cumulative residual entropy (CRE) and call it generalized cumulative entropy. The proposed entropy not only retains attributes of the existing uncertainty measures but also possesses the absolute homogeneous property with unbounded support, which the CRE does not have. We demonstrate its mathematical properties including the entropy of order statistics and the principle of maximum general cumulative entropy. We also introduce the cumulative ratio information as a measure of discrepancy between two distributions and examine its application to a goodness-of-fit test of the logistic distribution. Simulation study shows that the test statistics based on the cumulative ratio information have comparable statistical power with competing test statistics.     

On weighted generalized entropy

In the literature of information theory, the concept of generalized entropy has been proposed and the length-based shift dependent information measure has been studied. In this paper, the concept of weighted generalized entropy has been introduced. The properties of weighted generalized residual entropy and weighted generalized past entropy are also discussed.     

The generalized maximum Tsallis entropy estimators and applications to the Portland cement dataset

Tsallis entropy is a generalized form of entropy and tends to be Shannon entropy when q → 1. Using Tsallis entropy, an alternative estimation methodology (generalized maximum Tsallis entropy) is introduced and used to estimate the parameters in a linear regression model when the basic data are ill-conditioned. We describe the generalized maximum Tsallis entropy and for q = 2 we call that GMET2 estimator. We apply the GMET2 estimator for estimating the linear regression model Y = Xβ + e where the design matrix X is subject to severe multicollinearity. We compared the GMET2, generalized maximum entropy (GME), ordinary least-square (OLS), and inequality restricted least-square (IRLS) estimators on the analyzed dataset on Portland cement.     

Weighted extensions of generalized cumulative residual entropy and their applications

Abstract

In this paper, we consider weighted extensions of generalized cumulative residual entropy and its dynamic(residual) version. Our results include linear transformations, stochastic ordering, bounds, aging class properties and some relationships with other reliability concepts. We also define the conditional weighted generalized cumulative residual entropy and discuss some properties of its. For these concepts, we obtain some characterization results under some assumptions. Finally, we provide an estimator of the new information measure using empirical approach. In addition, we study large sample properties of this estimator.     

Characterizations based on generalized cumulative residual entropy functions

The Shannon entropy and the cumulative residual entropy (CRE) of a random variable are useful tools in probability theory. Recently, a new concept called generalized cumulative residual entropy (GCRE) of order n was introduced and studied. It is related with the record values of a sequence of i.i.d. random variables and with the relevation transform. In this paper, we show that, under some assumptions, the GCRE function of a fixed order n uniquely determines the distribution function. Some characterizations of particular probability models are obtained from this general result.     

A Fixed Point Approach to the Classification of Markov Chains with a Tree Structure

In this paper, we study the classification problem of discrete time and continuous time Markov processes with a tree structure. We first show some useful properties associated with the fixed points of a nondecreasing mapping. Mainly we find the conditions for a fixed point to be the minimal fixed point by using fixed point theory and degree theory. We then use these results to identify conditions for Markov chains of M/G/1 type or GI/M/1 type with a tree structure to be positive recurrent, null recurrent, or transient. The results are generalized to Markov chains of matrix M/G/1 type with a tree structure. For all these cases, a relationship between a certain fixed point, the matrix of partial differentiation (Jacobian) associated with the fixed point, and the classification of the Markov chain with a tree structure is established. More specifically, we show that the Perron-Frobenius eigenvalue of the matrix of partial differentiation associated with a certain fixed point provides information for a complete classification of the Markov chains of interest.     

Bivariate extension of generalized cumulative past entropy

The cumulative past entropy (CPE) of order α, a dual measure of cumulative residual entropy (CRE) of order α, has recently been proposed as a suitable extension of CPE. In this article, we extend the definition of (dynamic) CPE of order α (DCPE(α)) to bivariate setup and obtain some of its properties including bounds. We also look into the problem of extending DCPE(α) for conditionally specified models. Several properties, including monotonicity, and bounds of DCPE(α) are obtained for conditional distributions. Along with some characterization results it is shown that the proposed generalized measure uniquely determines the distribution function. Moreover, we also propose a stochastic order based on this measure and prove interrelation with some existing stochastic orders.     

Data disaggregation procedures within a maximum entropy framework

The aim of this paper is to formulate an analytical–informational–theoretical approach which, given the incomplete nature of the available micro-level data, can be used to provide disaggregated values of a given variable. A functional relationship between the variable to be disaggregated and the available variables/indicators at the area level is specified through a combination of different macro- and micro-data sources. Data disaggregation is accomplished by considering two different cases. In the first case, sub-area level information on the variable of interest is available, and a generalized maximum entropy approach is employed to estimate the optimal disaggregate model. In the second case, we assume that the sub-area level information is partial and/or incomplete, and we estimate the model on a smaller scale by developing a generalized cross-entropy-based formulation. The proposed spatial-disaggregation approach is used in relation to an Italian data set in order to compute the value-added per manufacturing sector of local labour systems within the Umbria region, by combining the available micro/macro-level data and by formulating a suitable set of constraints for the optimization problem in the presence of errors in micro-aggregates.     

On generalized dynamic cumulative past entropy measure

Recently, the concept of dynamic cumulative residual entropy and its generalizations has gained much attention among researchers. In this work, a new generalized dynamic cumulative measure in the past lifetime is proposed. Further, some characterization results connecting this new generalized dynamic entropy measure and other reversed measures are obtained.     

A quantile-based generalized dynamic cumulative measure of entropy

The cumulative residual entropy (CRE), introduced by Rao et al. (2004), is a new measure of uncertainty and viewed as a dynamic measure of uncertainty. Asadi and Zohrevand (2007) proposed a dynamic form of the CRE, namely dynamic CRE. Recently, Kumar and Taneja (2011) introduced a generalized dynamic CRE based on the Varma entropy introduced by Varma (1966) and called it dynamic CRE of order α and type β. In the present article, we introduce a quantile version of the dynamic CRE of order α and type β and study its properties. For this measure, we obtain some characterization results, aging classes properties, and stochastic comparisons.     

Contribution of maximum entropy principle in the field of queueing theory

ABSTRACT

In the literature of information theory, there exist many well known measures of entropy suitable for entropy optimization principles towards applications in different disciplines of science and technology. The object of this article is to develop a new generalized measure of entropy and to establish the relation between entropy and queueing theory. To fulfill our aim, we have made use of maximum entropy principle which provides the most uncertain probability distribution subject to some constraints expressed by mean values.     

Some new results on the cumulative residual entropy

The residual entropy function is a relevant dynamic measure of uncertainty in reliability and survival studies. Recently, Rao et al. [2004. Cumulative residual entropy: a new measure of information. IEEE Transactions on Information Theory 50, 1220–1228] and Asadi and Zohrevand [2007. On the dynamic cumulative residual entropy. Journal of Statistical Planning and Inference 137, 1931–1941] define the cumulative residual entropy and the dynamic cumulative residual entropy, respectively, as some new measures of uncertainty. They study some properties and applications of these measures showing how the cumulative residual entropy and the dynamic cumulative residual entropy are connected with the mean residual life function. In this paper, we obtain some new results on these functions. We also define and study the dynamic cumulative past entropy function. Some results are given connecting these measures of a lifetime distribution and that of the associated weighted distribution.     

On the dynamic cumulative residual entropy

Recently, Rao et al. [(2004) Cumulative residual entropy: a new measure of information. IEEE Trans. Inform. Theory 50(6), 1220–1228] have proposed a new measure of uncertainty, called cumulative residual entropy (CRE), in a distribution function F and obtained some properties and applications of that. In the present paper, we propose a dynamic form of CRE and obtain some of its properties. We show how CRE (and its dynamic version) is connected with well-known reliability measures such as the mean residual life time.     

Robust methods of estimation of regression coefficients 1

Recently, cumulative residual entropy (CRE) has been found to be a new measure of information that parallels Shannon's entropy (see Rao et al. [Cumulative residual entropy: A new measure of information, IEEE Trans. Inform. Theory. 50(6) (2004), pp. 1220–1228] and Asadi and Zohrevand [On the dynamic cumulative residual entropy, J. Stat. Plann. Inference 137 (2007), pp. 1931–1941]). Motivated by this finding, in this paper, we introduce a generalized measure of it, namely cumulative residual Renyi's entropy, and study its properties. We also examine it in relation to some applied problems such as weighted and equilibrium models. Finally, we extend this measure into the bivariate set-up and prove certain characterizing relationships to identify different bivariate lifetime models.     

Dependent models for observations which include angular ones

This paper discusses some stochastic models for dependence of observations which include angular ones. First, we provide a theorem which constructs four-dimensional distributions with specified bivariate marginals on certain manifolds such as two tori, cylinders or discs. Some properties of the submodel of the proposed models are investigated. The theorem is also applicable to the construction of a related Markov process, models for incomplete observations, and distributions with specified marginals on the disc. Second, two maximum entropy distributions on the cylinder are discussed. The circular marginal of each model is distributed as the generalized von Mises distribution which represents a symmetric or asymmetric, unimodal or bimodal shape. The proposed cylindrical model is applied to two data sets.     

Characterization based on generalized entropy of order statistics

ABSTRACT

In the present study, several characterizations of order statistics are obtained on the basis of the generalized entropy. Under some conditions, it is shown that the parent distribution can be uniquely determined by equality of generalized entropy of order statistics.     

Maximum entropy simulation for microdata protection

The paper proposes a new disclosure limitation procedure based on simulation. The key feature of the proposal is to protect actual microdata by drawing artificial units from a probability model, that is estimated from the observed data. Such a model is designed to maintain selected characteristics of the empirical distribution, thus providing a partial representation of the latter. The characteristics we focus on are the expected values of a set of functions; these are constrained to be equal to their corresponding sample averages; the simulated data, then, reproduce on average the sample characteristics. If the set of constraints covers the parameters of interest of a user, information loss is controlled for, while, as the model does not preserve individual values, re-identification attempts are impaired-synthetic individuals correspond to actual respondents with very low probability.Disclosure is mainly discussed from the viewpoint of record re-identification. According to this definition, as the pledge for confidentiality only involves the actual respondents, release of synthetic units should in principle rule out the concern for confidentiality.The simulation model is built on the Italian sample from the Community Innovation Survey (CIS). The approach can be applied in more generality, and especially suits quantitative traits. The model has a semi-parametric component, based on the maximum entropy principle, and, here, a parametric component, based on regression. The maximum entropy principle is exploited to match data traits; moreover, entropy measures uncertainty of a distribution: its maximisation leads to a distribution which is consistent with the given information but is maximally noncommittal with regard to missing information.Application results reveal that the fixed characteristics are sustained, and other features such as marginal distributions are well represented. Model specification is clearly a major point; related issues are selection of characteristics, goodness of fit and strength of dependence relations.