Embedding a state space model into a Markov decision process

In agriculture Markov decision processes (MDPs) with finite state and action space are often used to model sequential decision making over time. For instance, states in the process represent possible levels of traits of the animal and transition probabilities are based on biological models estimated from data collected from the animal or herd.     

Experimental study of a Bayesian method for daily electricity load forecasting

The method of Bayesian model discrimination is investigated for the possible contributions it may provide in the area of automatically forecasting the daily electricity demand cycle. A set of differing demand models have probabilities attached to them in such a way that these would be continuously updated with the available data and the actual forecasts obtained as expectations across all the models. Simulation experiments indicate significantly improved forecasting performance over a commonly used rescaling type of approach. Some practical issues in implementation are discussed.     

Conservatism of the z Confidence Interval Under Symmetric and Asymmetric Departures from Normality

The robustness of the z-interval under nonnormality is investigated by finding its infimum coverage probability over suitably chosen broad class of distributions. In the case of n = 1, the infimum coverage probabilities over the normal scale mixture and the symmetric unimodal families of distributions are obtained analytically. For general n, the infimum problem is theoretically reduced to a finite dimensional minimization which is then obtained numerically. The obtained minimum coverage probabilities are very close to the nominal probabilities. These exact minimum coverage probabilities are often notably sharper than the lower bounds given by the Camp-Meidell-Gauss inequality. The family of general unimodal distributions is considered next to investigate the possible effect of asymmetry. The obtained infimum coverage probabilities over this family are found to be the same as the ones over the symmetric unimodal class.     

Exponential Bounds for Ruin Probability in Two Moving Average Risk Models with Constant Interest Rate

The authors consider two discrete-time insurance risk models. Two moving average risk models are introduced to model the surplus process, and the probabilities of ruin are examined in models with a constant interest force. Exponential bounds for ruin probabilities of an infinite time horizon are derived by the martingale method.     

Application of a Tauberian theorem to finite model theory

An extension of a Tauberian theorem of Hardy and Littlewood is proved. It is used to show that, for classes of finite models satisfying certain combinatorial and growth properties, Cesàro probabilities (limits of average probabilities over second order sentences) exist. Examples of such classes include the class of unary functions and the class of partial unary functions. It is conjectured that the result holds for the usual notion of asymptotic probability as well as Cesàro probability. Evidence in support of the conjecture is presented.     

A Bayesian Approach to the Linear Combination of Forecasts

Bayesian methodology is suggested as a valid approach to the combination of forecasts and a simple subjectivist procedure is presented. It is shown how subjective probabilities can be meaningfully assigned over a set of forecasting models and updated, according to a Bayesian process, when the forecast realizations become known.     

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.     

The use of Markov operators to constructing generalised probabilities

A new approach to constructing generalised probabilities is proposed. It is based on the models using lower and upper previsions, or equivalently, convex sets of probability measures. Our approach uses sets of Markov operators in the role of rules preserving desirability of gambles. The main motivation being the operators of conditional expectations which are usually assumed to reduce riskiness of gambles. Imprecise probability models are then obtained in the ways to be consistent with those desirability preserving rules. The consistency criteria are based on the existing interpretations of models using imprecise probabilities. The classical models based on lower and upper previsions are shown to be a special class of the generalised models. Further, we generalise some standard extension procedures, including the marginal extension and independent products, which can be defined independently of the existing procedures known for standard models.     

Decision making with interval probabilities

Handling uncertainty by interval probabilities is recently receiving considerable attention by researchers. Interval probabilities are used when it is difficult to characterize the uncertainty by point-valued probabilities due to partially known information. Most of researches related to interval probabilities, such as combination, marginalization, condition, Bayesian inferences and decision, assume that interval probabilities are known. How to elicit interval probabilities from subjective judgment is a basic and important problem for the applications of interval probability theory and till now a computational challenge. In this work, the models for estimating and combining interval probabilities are proposed as linear and quadratic programming problems, which can be easily solved. The concepts including interval probabilities, interval entropy, interval expectation, interval variance, interval moment, and the decision criteria with interval probabilities are addressed. A numerical example of newsvendor problem is employed to illustrate our approach. The analysis results show that the proposed methods provide a novel and effective alternative for decision making when point-valued subjective probabilities are inapplicable due to partially known information.     

How does innovation's tail risk determine marginal tail risk of a stationary financial time series?

We discuss the relationship between the marginal tail risk probability and theinnovation's tail risk probability for some stationary financial time series models. We firstgive the main results on the tail behavior of a class of infinite weighted sums of randomvariables with heavy-tailed probabilities. And then, the main results are applied to threeimportant types of time series models; infinite order moving averages, the simple bilineartime series and the solutions of stochastic difference equations. The explicit formulasare given to describe how the marginal tail probabilities come from the innovation's tailprobabilities for these time series. Our results can be applied to the tail estimation of timeseries and are useful for risk analysis in finance.     

Discrete time Markov chains with interval probabilities

The parameters of Markov chain models are often not known precisely. Instead of ignoring this problem, a better way to cope with it is to incorporate the imprecision into the models. This has become possible with the development of models of imprecise probabilities, such as the interval probability model. In this paper we discuss some modelling approaches which range from simple probability intervals to the general interval probability models and further to the models allowing completely general convex sets of probabilities. The basic idea is that precisely known initial distributions and transition matrices are replaced by imprecise ones, which effectively means that sets of possible candidates are considered. Consequently, sets of possible results are obtained and represented using similar imprecise probability models.We first set up the model and then show how to perform calculations of the distributions corresponding to the consecutive steps of a Markov chain. We present several approaches to such calculations and compare them with respect to the accuracy of the results. Next we consider a generalisation of the concept of regularity and study the convergence of regular imprecise Markov chains. We also give some numerical examples to compare different approaches to calculations of the sets of probabilities.     

Consistency of random field specifications

Summary In the random field approach to lattice gas models it has been shown that the one point conditional probabilities determine the finite set conditional probabilities under conditions of strict positivity and regularity. This paper considers the case when strict positivity does not obtain with families of conditional probabilities more general than the one-point conditional probabilities.     

Towards a constructive approach to act-conditional subjective expected utility models

In structuring a decision problem under uncertainty, the uncertain environment may be affected by the choice of an act. In decision analysis, the decision maker provides subjective probabilities and utilities through separate elicitation processes, and then both components are combined together to give an index of his preference over decision alternatives. Based upon this conceptualisation of decision analysis, a constructive approach to act-conditional subjective expected utility theory is proposed. Two utility models have been addressed: the linear utility model and the weighted utility model.     

Modeling mixed push and pull promotion flows in Manpower Planning

Manpower Planning is a useful tool for human resource management in large organizations. Classical Manpower Planning models are analytical time-discrete push and pull models. Push models are characterized by the same promotion and wastage probabilities for people within the same group. This assumption is suitable in organizations where for instance promotions are used for reasons of personnel motivation or employees are promoted after succeeding in an exam. In many organizations, people are only promoted when there are vacancies at other levels. In those cases, pull models can be used. Pull models only assume known wastage probabilities. In practice, both assumptions may occur simultaneously. In this paper, a mixed push-pull model is developed for organizations in which both types of flows are considered.     

Intensity models and transition probabilities for credit card loan delinquencies

We estimate the probability of delinquency and default for a sample of credit card loans using intensity models, via semi-parametric multiplicative hazard models with time-varying covariates. It is the first time these models, previously applied for the estimation of rating transitions, are used on retail loans. Four states are defined in this non-homogenous Markov chain: up-to-date, one month in arrears, two months in arrears, and default; where transitions between states are affected by individual characteristics of the debtor at application and their repayment behaviour since. These intensity estimations allow for insights into the factors that affect movements towards (and recovery from) delinquency, and into default (or not). Results indicate that different types of debtors behave differently while in different states. The probabilities estimated for each type of transition are then used to make out-of-sample predictions over a specified period of time.     

Conditionally Specified Space-Time Models for Multivariate Processes

This article proposes a class of conditionally specified models for the analysis of multivariate space-time processes. Such models are useful in situations where there is sparse spatial coverage of one of the processes and much more dense coverage of the other process(es). The dependence structure across processes and over space, and time is completely specified through a neighborhood structure. These models are applicable to both point and block sources; for example, multiple pollutant monitors (point sources) or several county-level exposures (block sources). We introduce several computational tricks that are integral for model fitting, give some simple sufficient and necessary conditions for the space-time covariance matrix to be positive definite, and implement a Gibbs sampler, using Hybrid MC steps, to sample from the posterior distribution of the parameters. Model fit is assessed via the DIC. Predictive accuracy, over both time and space, is assessed both relatively and absolutely via mean squared prediction error and coverage probabilities. As an illustration of these models, we fit them to particulate matter and ozone data collected in the Los Angeles, CA, area in 1995 over a three-month period. In these data, the spatial coverage of particulate matter was sparse relative to that of ozone.     

Estimating Mixture of Dirichlet Process Models

Abstract

Current Gibbs sampling schemes in mixture of Dirichlet process (MDP) models are restricted to using “conjugate” base measures that allow analytic evaluation of the transition probabilities when resampling configurations, or alternatively need to rely on approximate numeric evaluations of some transition probabilities. Implementation of Gibbs sampling in more general MDP models is an open and important problem because most applications call for the use of nonconjugate base measures. In this article we propose a conceptual framework for computational strategies. This framework provides a perspective on current methods, facilitates comparisons between them, and leads to several new methods that expand the scope of MDP models to nonconjugate situations. We discuss one in detail. The basic strategy is based on expanding the parameter vector, and is applicable for MDP models with arbitrary base measure and likelihood. Strategies are also presented for the important class of normal-normal MDP models and for problems with fixed or few hyperparameters. The proposed algorithms are easily implemented and illustrated with an application.     

Fuzzy probability vectors

Several recent papers have discussed an extension of decision analysis: both the rewards and probabilities are modelled using fuzzy sets. However, arbitrary specification of the membership functions of the fuzzy probabilities in these models may lead to internal inconsistencies. To avoid this, the fuzzy beta possibility distribution is introduced.     

一类搜索型随机格斗的获胜概率研究

把随机过程分析引入Lanchester方程就形成了随机格斗理论.运用随机格斗理论研究了潜艇协同隐蔽攻击水面舰艇编队获胜概率的数学模型,利用状态转移图和Laplace变换的性质推导出了2对2搜索型随机格斗中双方的获胜概率公式,并结合潜艇协同隐蔽攻击水面舰艇的实际,计算分析了格斗双方的获胜概率.利用这一公式可以得到概率上的精确解,能够被用来定量评估潜艇协同隐蔽攻击水面舰艇编队的作战效能.     

Dependence properties and bounds for ruin probabilities in multivariate compound risk models

In risk management, ignoring the dependence among various types of claims often results in over-estimating or under-estimating the ruin probabilities of a portfolio. This paper focuses on three commonly used ruin probabilities in multivariate compound risk models, and using the comparison methods shows how some ruin probabilities increase, whereas the others decrease, as the claim dependence grows. The paper also presents some computable bounds for these ruin probabilities, which can be calculated explicitly for multivariate phase-type distributed claims, and illustrates the performance of these bounds for the multivariate compound Poisson risk models with slightly or highly dependent Marshall-Olkin exponential claim sizes.