Thurstone scaling in order statistics

Thurstone scaling is widely used for presenting the priorities among the compared items. The mean values of the quantiles corresponding to frequencies of each stimilus’ preference over the other stimuli define the items’ locations on the psychological continuum of the Thurstone scale. This paper considers an extension of the scale levels to the aggregates of the independent covariates. In a sense, it is similar to a multiple regression extension of the mean value of the dependent variable to its conditional mean expressed by the linear aggregate of the independent variables. A maximum likelihood objective constructed by the probabilities of the order statistics applied to the ranked or paired comparison data is suggested. Probit, logit and multinomial links are tried to obtain the Thurstonian scale exposition by the covariates, and to estimate probabilities of the items’ choice. This approach is very convenient and can substantially enrich both theoretical interpretation and practical application of Thurstone modelling, particularly in marketing and advertising research.     

Natural atomic probabilities in quantum information theory

Quantum Information Theory has witnessed a great deal of interest in the recent years since its potential for allowing the possibility of quantum computation through quantum mechanics concepts such as entanglement, teleportation and cryptography. In Chemistry and Physics, von Neumann entropies may provide convenient measures for studying quantum and classical correlations in atoms and molecules. Besides, entropic measures in Hilbert space constitute a very useful tool in contrast with the ones in real space representation since they can be easily calculated for large systems. In this work, we show properties of natural atomic probabilities of a first reduced density matrix that are based on information theory principles which assure rotational invariance, positivity, and N- and v-representability in the Atoms in Molecules (AIM) scheme. These (natural atomic orbital-based) probabilities allow the use of concepts such as relative, conditional, mutual, joint and non-common information entropies, to analyze physical and chemical phenomena between atoms or fragments in quantum systems with no additional computational cost. We provide with illustrative examples of the use of this type of atomic information probabilities in chemical process and systems.     

Independence for full conditional probabilities: Structure,factorization, non-uniqueness,and Bayesian networks

This paper examines concepts of independence for full conditional probabilities; that is, for set-functions that encode conditional probabilities as primary objects, and that allow conditioning on events of probability zero. Full conditional probabilities have been used in economics, in philosophy, in statistics, in artificial intelligence. This paper characterizes the structure of full conditional probabilities under various concepts of independence; limitations of existing concepts are examined with respect to the theory of Bayesian networks. The concept of layer independence (factorization across layers) is introduced; this seems to be the first concept of independence for full conditional probabilities that satisfies the graphoid properties of Symmetry, Redundancy, Decomposition, Weak Union, and Contraction. A theory of Bayesian networks is proposed where full conditional probabilities are encoded using infinitesimals, with a brief discussion of hyperreal full conditional probabilities.     

Sensitivity analysis of discrete Markov chains via matrix calculus

Markov chains are often used as mathematical models of natural phenomena, with transition probabilities defined in terms of parameters that are of interest in the scientific question at hand. Sensitivity analysis is an important way to quantify the effects of changes in these parameters on the behavior of the chain. Many properties of Markov chains can be written as simple matrix expressions, and hence matrix calculus is a powerful approach to sensitivity analysis. Using matrix calculus, we derive the sensitivity and elasticity of a variety of properties of absorbing and ergodic finite-state chains. For absorbing chains, we present the sensitivities of the moments of the number of visits to each transient state, the moments of the time to absorption, the mean number of states visited before absorption, the quasistationary distribution, and the probabilities of absorption in each of several absorbing states. For ergodic chains, we present the sensitivity of the stationary distribution, the mean first passage time matrix, the fundamental matrix, and the Kemeny constant. We include two examples of application of the results to demographic and ecological problems.     

Estimating Invariant Probability Densities for Dynamical Systems

Knowing a probability density (ideally, an invariant density) for the trajectories of a dynamical system allows many significant estimates to be made, from the well-known dynamical invariants such as Lyapunov exponents and mutual information to conditional probabilities which are potentially more suitable for prediction than the single number produced by most predictors. Densities on typical attractors have properties, such as singularity with respect to Lebesgue measure, which make standard density estimators less useful than one would hope. In this paper we present a new method of estimating densities which can smooth in a way that tends to preserve fractal structure down to some level, and that also maintains invariance. We demonstrate with applications to real and artificial data.     

Control problems with time-lag. II

Consider a continuous time Markov chain with stationary transition probabilities. A function of the state is observed. A regular conditional probability distribution for the trajectory of the chain, given observations up to time t, is obtained. This distribution also corresponds to a Markov chain, but the conditional chain has nonstationary transition probabilities. In particular, computation of the conditional distribution of the state at time s is discussed. For s > t, we have prediction (extrapolation), while s < t corresponds to smoothing (interpolation). Equations for the conditional state distribution are given on matrix form and as recursive differential equations with varying s or t. These differential equations are closely related to Kolmogorov's forward and backward equations. Markov chains with one observed and one unobserved component are treated as a special case. In an example, the conditional distribution of the change-point is derived for a Poisson process with a changing intensity, given observations of the Poisson process.     

Tail dependence of random variables from ARCH and heavy-tailed bilinear models

Discussed in this paper is the dependent structure in the tails of distributions of random variables from some heavy-tailed stationary nonlinear time series. One class of models discussed is the first-order autoregressive conditional heteroscedastic (ARCH) process introduced by Engle (1982). The other class is the simple first-order bilinear models driven by heavy-tailed innovations. We give some explicit formulas for the asymptotic values of conditional probabilities used for measuring the tail dependence between two random variables from these models. Our results have significant meanings in finance.     

Orthant tail dependence of multivariate extreme value distributions

The orthant tail dependence describes the relative deviation of upper- (or lower-) orthant tail probabilities of a random vector from similar orthant tail probabilities of a subset of its components, and can be used in the study of dependence among extreme values. Using the conditional approach, this paper examines the extremal dependence properties of multivariate extreme value distributions and their scale mixtures, and derives the explicit expressions of orthant tail dependence parameters for these distributions. Properties of the tail dependence parameters, including their relations with other extremal dependence measures used in the literature, are discussed. Various examples involving multivariate exponential, multivariate logistic distributions and copulas of Archimedean type are presented to illustrate the results.     

Invariance of stationary probabilities of states for network of multiserver queues

A queueing network consisting of multiserver nodes with different sources of customers is considered. We investigate stationary probability distributions of network states. Their invariance with respect to the functional form of distributions of input and service processes is established under fixed expectations of these distributions and service discipline when any entering customer has a preemptive resume discipline.     

The L-separation criterion for description of cs-independence models

Different conditional independence models have been proposed in literature; in this paper we consider models induced by conditional probabilities based on the definition of conditional cs-independence. These models need not comply with the symmetry property, so that they have not the graphoid structure. Hence, the well-known d-separation criterion for directed acyclic graphs may not be able to represent such independence models. Therefore, we introduce a new separation criterion called L-separation. We study its main properties and show how it allows to represent the above-mentioned independence models through directed acyclic graphs.     

A Coherent Aggregation Framework for Stress Testing and Scenario Analysis

Abstract

We present a methodology to aggregate in a coherent manner conditional stress losses in a trading or banking book. The approach bypasses the specification of unconditional probabilities of the individual stress events and ensures by a linear programming approach so that the (subjective or frequentist) conditional probabilities chosen by the risk manager are internally consistent. The admissibility requirement greatly reduces the degree of arbitrariness in the conditional probability matrix if this is assigned subjectively. The approach can be used to address the requirements of the regulators on the Instantaneous Risk Charge.     

Simultaneous Posterior Probability Statements From Monte Carlo Output

This article considers the problem of making simultaneous probability statements in multivariate inferential problems based on samples from a posterior distribution. The calculation of simultaneous credible bands is reviewed and—as an alternative—contour probabilities are proposed. These are defined as 1 minus the content of the highest posterior density region which just covers a certain point of interest. We discuss a Monte Carlo method to estimate contour probabilities and distinguish whether or not the functional form of the posterior density is available. In the latter case, an approach based on Rao-Blackwellization is proposed. We highlight that this new estimate has an important invariance property. We illustrate the performance of the different methods in three applications.     

Canonical and microcanonical Gibbs states

Summary The grand canonical Gibbs states for a system from classical statistical mechanics can be defined as the probability measures on an appropriate phase space which have certain specified conditional probabilities. These conditional probabilities are with respect to a family of -algebras associated with subsets of the space in which the system lies. If different families of -algebras are used then canonical and microcanonical Gibbs states are obtained. The relationship between these different Gibbs states is studied and, subject to various conditions, it is shown that each canonical and microcanonical Gibbs state can be written as a convex mixture of grand canonical Gibbs states.     

On the Weak Invariance Principle for Stationary Sequences under Projective Criteria

In this paper, we study the central limit theorem and its weak invariance principle for sums of a stationary sequence of random variables, via a martingale decomposition. Our conditions involve the conditional expectation of sums of random variables with respect to the distant past. The results contribute to the clarification of the central limit question for stationary sequences. Magda Peligrad is supported in part by a Charles Phelps Taft research support grant at the Univeristy of Cincinnati and the NSA grant H98230-05-1-0066.     

Propagation effects of model-calculated probability values in Bayesian networks

Probabilistic causal interaction models have become quite popular among Bayesian-network engineers as elicitation of all probabilities required often proves the main bottleneck in building a real-world network with domain experts. The best-known interaction models are the noisy-OR model and its generalisations. These models in essence are parameterised conditional probability tables for which just a limited number of parameter probabilities are required. The models assume specific properties of intercausal interaction and cannot be applied uncritically. Given their clear engineering advantages however, they are subject to ill-considered use. This paper demonstrates that such ill-considered use can result in poorly calibrated output probabilities from a Bayesian network. By studying, in an analytical way, the propagation effects of noisy-OR calculated probability values, we identify conditions under which use of the model can be harmful for a network's performance. These conditions demonstrate that use of the noisy-OR model for mere pragmatic reasons is sometimes warranted, even when the model's underlying assumptions are not met in reality.     

Duality in non-additive expected utility theory

Some duality problems in expected utility theory, raised by the introduction of non-additive probabilities, are examined. These problems do not arise if the probability measure is symmetric; i.e. has the property of complementary additivity. Additional, mild properties of coherence of conditional probabilities imply full additivity of the unconditional measure.     

Generalized symmetrized dirichlet distributions

In this paper we introduce a family of multivariate distributions, which consists of scale mixtures of symmetrized Dirichlet distributions. This family is a symmetrization of multivariate Liouville distributions and contains the well-known spherically symmetric distributions as a special case. The basic properties of this family such as stochastic representation, probability density functions, marginal and conditional distributions and components' independence are studied. A criterion of the invariance of statistics is also given.     

Conditional possibilities independence and noninteraction

Relations from the theory of probability are compared with corresponding relations in the theory of possibility. Because of the special properties of the min operator, the values of the conditional possibilities are much more similar to those of the joint possibilities than is the case in connection with probabilities. When the joint possibilities are given, the solution for the conditional possibilities is not always unique. It is shown that we must distinguish between noninteraction and independence of fuzzy or possibilistic variables. For purposes of inference, relations representing conditional possibility distributions should be used.     

EPSTA: The coincidence of time-stationary and customer-stationary distributions

In the first part of this paper we present an overview of relationships between time- and customer-stationary distributions of queueing processes. These have been proved by using the properties of random marked point processes, stochastic processes with embedded point processes, Palm distributions and an intensity conservation principle. In the second part a necessary and sufficient condition is established for the coincidence of the two types of stationary distributions, using conditional intensities. We also formulate the property of EPSTA that includes PASTA and ASTA as particular cases. A further result concerns the conditional EPSTA property. Applications to particular queueing systems are considered.